Design, synthesis, and biological evaluation of Erythrina alkaloid analogues as neuronal nicotinic acetylcholine receptor antagonists

Article abstract of DOI:10.1021/jm4013592

The synthesis of a new series of Erythrina alkaloid analogues and their pharmacological characterization at various nicotine acetylcholine receptor (nAChR) subtypes are described. The compounds were designed to be simplified analogues of aromatic erythrinanes with the aim of obtaining subtype-selective antagonists for the nAChRs and thereby probe the potential of using these natural products as scaffolds for further ligand optimization. The most selective and potent nAChR ligand to come from the series, 6,7-dimethoxy-2- methyl-1,2,3,4-tetrahydroisoquinoline (3c) (also a natural product by the name of O-methylcorypalline), displayed submicromolar binding affinity toward the α4β2 nAChR with more than 300-fold selectivity over α4β4, α3β4, and α7. Furthermore, this lead structure (which also has inhibitory activity at monoamine oxidases A and B and at the serotonin and norepinephrine transporters) showed antidepressant-like effect in the mouse forced swim test at 30 mg/kg.

Full text of DOI:10.1021/jm4013592

Article
pubs.acs.org/jmc
Design, Synthesis, and Biological Evaluation of Erythrina Alkaloid
Analogues as Neuronal Nicotinic Acetylcholine Receptor Antagonists
Franco̧
is Crestey,†,§ Anders A. Jensen,†,∥ Morten Borch,† Jesper Tobias Andreasen,† Jacob Andersen,†
Thomas Balle,‡ and Jesper Langgaard Kristensen*,†
†Department of Drug Design and Pharmacology, Faculty of Health and Medical Sciences, University of Copenhagen,
Universitetsparken 2, 2100 Copenhagen, Denmark
‡Faculty of Pharmacy, University of Sydney, Sydney, New South Wales 2006, Australia
S
* Supporting Information
ABSTRACT: The synthesis of a new series of Erythrina alkaloid analogues and their pharmacological characterization at various
nicotine acetylcholine receptor (nAChR) subtypes are described. The compounds were designed to be simplified analogues of
aromatic erythrinanes with the aim of obtaining subtype-selective antagonists for the nAChRs and thereby probe the potential of
using these natural products as scaffolds for further ligand optimization. The most selective and potent nAChR ligand to come
from the series, 6,7-dimethoxy-2-methyl-1,2,3,4-tetrahydroisoquinoline (3c) (also a natural product by the name of O-
methylcorypalline), displayed submicromolar binding affinity toward the α4β2 nAChR with more than 300-fold selectivity over
α4β4, α3β4, and α7. Furthermore, this lead structure (which also has inhibitory activity at monoamine oxidases A and B and at
the serotonin and norepinephrine transporters) showed antidepressant-like effect in the mouse forced swim test at 30 mg/kg.
erysopine, erysodine, erysovine, and erythraline (Figure 1). The
activities of some of these compounds, erysotrine, erysopine,
erysodine, and also DHβE, have been investigated at the
nAChRs.5 Erysodine was found to be the most potent
competitive antagonist of the four at the α4β2 nAChR,
displaying a binding affinity of 50 nM in a [3H]cytisine binding
assay and a Ki value of 7.5 μM at the α7 nAChR.6,7 Moreover,
erysodine has also been compared with DHβE in vivo where it
prevented both the early developing decrease and the late-
developing increase in locomotor activity produced by (S)-
nicotine in rats, and it produced a greater separation between
the effective dose and the dose where toxic motor effects were
observed.8 The latter observation can presumably be ascribed
to erysodine having a greater selectivity for neuronal nAChR
over the muscle-type receptor than DHβE.
The potent antagonism exerted by erysodine together with
its ability to enter the brain after systemic administration makes
this natural product an interesting pharmacological tool for
studies of the nAChRs in vivo. The limited availability of
erysodine and analogues from natural sources precludes the
widespread applications of these ligands, but at the same time
this makes them very interesting lead structures in the pursuit
of simpler and more accessible analogues. Whereas previous
“deconstruction” of other natural product like MLA has
INTRODUCTION
■
The neuronal nicotinic acetylcholine receptors (nAChRs) have
been proposed to be involved in a number of different disease
states of the central nervous system (CNS), including
depression, schizophrenia, attention deficit hyperactivity
disorder, Alzheimer’s and Parkinson’s diseases, substance
abuse, and pain, and this has made the receptors popular
targets in drug discovery.1 A common feature of the vast
majority of nAChR compounds developed to date is that they
are either agonists or positive allosteric modulators of the
receptors.2 In contrast, much less attention has been directed
toward the development of antagonists targeting the neuronal
nAChRs.3 Most of the available selective nAChR antagonists
are natural products, and many of them, e.g., D-tubocurarine,
methyllycaconitine (MLA), and α-conotoxins, are not optimal
as lead compounds for CNS drug discovery due to their large
sizes and unfavorable physicochemical properties. A notable
exception is dihydro-β-erythroidine (DHβE). DHβE is a
member of the Erythrina alkaloid family characterized by
their unique tetracyclic spiroamine framework (Figure 1).4
Erythrinanes were first isolated from Erythrina species at the
end of the 19th century. Extracts from these plants were found
to have curare-like neuromuscular blocking effects, and
systematic phytochemical examination of the genus Erythrina
has identified more than 100 erythrinanes that can be
subdivided into several classes. Of these, the most abundant
classes are the aromatic erythrinanes, including, e.g., erysotrine,
Received: September 4, 2013
Published: November 4, 2013
© 2013 American Chemical Society
9673
dx.doi.org/10.1021/jm4013592 | J. Med. Chem. 2013, 56, 9673−9682

Journal of Medicinal Chemistry
Article
Figure 1. The erythrinane scaffold, DHβE, and structure of aromatic erythrinanes.
Figure 2. Aromatic erythrinanes and new ligands in which the structural complexity is gradually reduced.
a
Scheme 1. Synthesis of the Phenethylamines 4a−f
a
Reagents and conditions: (i) benzyl bromide (1.2 equiv), K2CO3 (1.5 equiv), EtOH, 70 °C, 97−98%; (ii) isopropyl bromide (1.2 equiv), K2CO3
(1.5 equiv), EtOH, 70 °C, 56−70%; (iii) NO2Me (3.6 equiv), NH4OAc (3 equiv), AcOH, 120 °C, 49−81%; (iv) LiAlH4 (5 equiv), dry THF, 0 °C
to reflux conditions, 84−99%.
a
Scheme 2. Synthesis of cis-Erythrinane Derivatives 1a−g
a
Reagents and conditions: (i) method A; (a) AlMe3 (2 equiv), ethyl 2-cyclohexanoneacetate (1.05 equiv), Sc(OTf)3 (12.5 mol %), CH3CN, 0 °C to
rt, (b) TfOH (3.5 equiv), 0 °C to rt; method B, (a) ethyl 2-cyclohexanoneacetate (1.05 equiv), PhMe, reflux conditions, (b) aq HI (57%), rt. (ii)
LiAlH4 (2.5 equiv), dry THF, 0 °C to rt, 20−56% over two steps. (iii) concd H2SO4 (8.8 equiv), AcOH, reflux conditions, 79−80%. (iv) BBr3 (3
equiv), DCM, 0 °C to rt, 64%.
provided some interesting small molecule nAChR ligands,9
reports on structure−activity relationships of the Erythrina
alkaloids on the nAChRs are very scarce.10
In the search for new competitive nAChR antagonists, we
embarked on the synthesis of ligands in which the structure of
the aromatic erythrinanes is simplified step by step in order to
identify the structural elements in the molecules essential for
their nAChR activity. Bermudez et al. have investigated the
binding mode of several erythrinanes to the nAChR using a
combination of mutational studies and molecular modeling, and
shown that the substitution pattern on the catecol unit has a
profound effect on the affinity of the different alkaloids.6 Thus,
we decided to focus our study on two parameters: number of
rings in the molecules and substitution pattern on the catecol
unit. Ease of synthesis was given priority so that the ligands
could be accessed rapidly. First, the methoxy group on the A-
ring was omitted and the double bonds in the A- and B-ring
were saturated, giving the first series of analogues, then the A-
9674
dx.doi.org/10.1021/jm4013592 | J. Med. Chem. 2013, 56, 9673−9682

Journal of Medicinal Chemistry
Article
a
Scheme 3. Synthesis of Tetrahydroisoquinoline Derivatives 2a−g
a
Reagents and conditions: (i) γ-butyrolactone (1.1 equiv), PhMe, reflux conditions, 46−72%; (ii) (a) POCl3 (10 equiv), PhMe, reflux conditions,
(b) NaBH4 (2 equiv), EtOH−AcOH (10:1), 0 °C to rt, 42−69%; (iii) H2, Pd/C (10 mol %), MeOH, rt, 72−75%; (iv) BBr3 (3 equiv), DCM, 0 °C
to rt, 83%.
and B-rings were deleted sequentially, giving the final two series
of analogues, see Figure 2. We included all possible
permutations of H, CH3, and −CH2− on the catechol unit
that is also present in the natural products, thus giving 15
ligands in the initial set of compounds. Following the
pharmacological characterization of these ligands, we synthe-
sized a further six ligands based on the most promising scaffold.
However, none of these ligands showed an improved profile.
Given the evidence for a role of nAChRs in depression and
preclinical studies demonstrating antidepressant-like action of
nAChR antagonists in rodents, the present study was finalized
by evaluating the pharmacological properties of the initial lead
compound at monoamine transporters and investigate its
possible antidepressant activity in the mouse forced swim test.
The racemic pyrrolo[2,1-a]isoquinoline derivatives 2a−g
were prepared in a two-step protocol described by Bailey and
co-workers as depicted in Scheme 3.17 Phenethylamines 4a−b
and 4e−f reacted with γ-butyrolactone, providing amides 8a−d
which cyclized upon treatment with POCl3. Subsequent
reduction of the intermediate iminium salts with NaBH4 gave
isoquinoline substrates 2a−d with 42−69% yields. Debenzyla-
tion of 2a and 2b using H2/Pd(c) in methanol gave tricyclic
analogues 2e and 2f in 72% and 75% yields, respectively.
Compound 2g was obtained from derivative 2c as a
hydrobromide by following the same procedure as previously
described above.
The N-methyl-1,2,3,4-tetrahydroisoquinolines 3a−d were
obtained in a single-step procedure by performing a Pictet−
Spengler reaction with the appropriate phenethylamines 4
using 5 equiv of formaldehyde in formic acid.18 The subsequent
debenzylation using the same conditions as described above
furnished 3e and 3f in 83% and 89% yields, respectively. The
treatment of dimethoxyisoquinoline 3c with BBr3 afforded
dihydroxyisoquinoline 3g as its hydrobromic salt in 69% yield
(Scheme 4).
CHEMISTRY
■
First, the synthesis of the required substituted phenethylamines
4a−f, common intermediates for the preparation of all the
designed ligands, is outlined in Scheme 1. Benzyl protection of
the hydroxyl group of commercially available isovanillin 5a and
vanillin 5b furnished aldehydes 6a and 6b in 98% and 97%
yields, respectively. The same conditions were applied for the
isopropyl protection reaction, giving aldehydes 6c and 6d in
good yields. The nitrostyrenes 7a−f were obtained from the
condensation of the corresponding protected aldehydes 6a−f
with nitromethane in 49−81% yields, and a subsequent
reduction with LiAlH4 yielded amines 4a−f.11
a
Scheme 4. Synthesis of Tetrahydroisoquinolines 3a−g
The syntheses of the cis-erythrinanes 1a−d have been
performed via two different pathways as shown in Scheme
2.12,13 Following a modified procedure (method A) recently
published by Tietze and co-workers,14 the spirocyclic core of
the Erythrina alkaloids was formed by a Lewis acid mediated
domino reaction of phenethylamines 4c and 4f with ethyl 2-
cyclohexanoneacetate, forming successively three new bonds in
one process. The resulting crude intermediates were reduced
with an excess of LiAlH4, providing erythrinane derivatives 1a
and 1d in 20% and 56% yields over two steps, respectively.
Tetracyclic aza-spiro analogues 1b and 1c were obtained from
phenethylamines 4d and 4e in moderate yields by applying the
procedure (method B) described by Isobe and co-workers,15
followed by the subsequent reduction with LiAlH4 of the crude
products. The deisopropylation of compounds 1a and 1b were
performed in AcOH/H2SO4 at reflux,16 leading to the desired
cis-erythrinane derivatives 1e and 1f. The treatment of 1c with
3 equivalents of BBr3 in dichloromethane afforded the
dihydroxy analogue 1g as a hydrobromide in 64% yield.
a
Reagents and conditions: (i) HCOH 37% aq (5 equiv), HCO2H, 100
°C, 58−99%; (ii) H2, Pd/C (10 mol %), MeOH, rt, 83−89%; (iii)
BBr3 (5 equiv), DCM, 0 °C to rt, 69%.
Subsequent to the pharmacological characterization of the 15
ligands detailed above, six analogues of 3c were synthesized
(9a−f) in which substituents were introduced at the 1-position
of the tetrahydroisoquinoline ring system. Indeed, treatment of
N-monomethylated amine 10 with the appropriate acyl
chlorides provided the corresponding amides which underwent
a Bischler−Napieralski cyclodehydration in the presence of
POCl3. Subsequent reduction of the iminium salts with NaBH4
led to 9a−f in 78−88% yield over three steps (Scheme 5).
9675
dx.doi.org/10.1021/jm4013592 | J. Med. Chem. 2013, 56, 9673−9682

Journal of Medicinal Chemistry
Article
the rat heteromeric nAChR subtypes α4β2, α3β4, and α4β4
and in a [3H]MLA binding assay using membranes from
tsA201 transiently expressing the α7/5-HT3A chimera (com-
posed of the amino-terminal domain of the rat α7 nAChR
subunit and the ion channel domain of the murine 5-HT3A
receptor subunit), essentially as previously described.19 The
functional characteristics of the analogues at a HEK293T cell
line stably expressing the mouse α4β2 nAChR and at a
HEK293 cell line stably expressing the rat α3β4 nAChR in the
FLIPR Membrane Potential Blue assay was determined as
previously described.20 (S)-Nicotine (EC70−EC80 for the
respective receptors, i.e., 1 μM for α4β2 and 3 μM for α3β4)
was used as agonist in the antagonist experiments.
Scheme 5. Synthesis of 1-Substituted
Tetrahydroisoquinolines 9a−f
a
a
Reagents and conditions: (i) (a) (Boc)2O (1.1 equiv), DCM, rt, (b)
LiAlH4 (1.5 equiv), dry THF, 0 °C to reflux conditions, 83% over 2
steps; (ii) (a) RCOCl (1.1 equiv), TEA (1.1 equiv), dry DCM, 0 °C to
rt, (b) POCl3 (10 equiv), dry PhMe, reflux conditions, (c) NaBH4 (5
equiv), MeOH, 0 °C to rt, 78−88% over 3 steps.
All 15 compounds in the series displaced [3H]epibatidine
binding from the rat α4β2 nAChR in a concentration-
dependent manner, exhibiting Ki values at the receptor ranging
from 0.55 to ∼100 μM. In contrast, the vast majority of the
compounds exhibited negligible binding affinity to the rat α4β4
and α3β4 nAChRs and to the α7/5-HT3A chimera (Table 1).
Overall, this SAR was also reflected in the inhibitory potencies
RESULTS AND DISCUSSION
■
The ligands were characterized in a [3H]epibatidine binding
assay using membranes from HEK293 cells stably expressing
a
Table 1. Chemical Stuctures of the Synthesized Compounds and Their Pharmacological Properties at nAChRs
binding Ki (μM)
functional IC50 (μM)
compd
R
R′
α4β2
∼100
α4β4
α3β4
α7/5-HT3A
α4β2
α3β4
b
b
b
1c
1d
1e
1f
CH3
−CH2−
H
CH3
∼100
∼50
∼50
23
∼100
∼30
∼100
∼30
∼30
18
31
∼50
∼50
∼50
7.3
9.2
2.3
2.9
CH3
H
8.5
20
CH3
H
18
5.5
33
b
b
b
b
b
b
b
b
b
1g
2c
2d
2e
2f
H
1.5
19
∼30
7.3
0.58
31
∼50
CH3
−CH2−
H
CH3
5.5
∼100
21
∼300
∼100
∼300
∼100
∼100
∼300
∼300
∼300
∼300
∼100
>500
∼100
∼500
∼300
∼300
>500
∼500
>500
>500
>500
∼100
∼100
∼300
∼300
∼1000
∼100
∼300
∼1000
>1000
>1000
0.55
3.3
b
CH3
H
∼100
3.4
11
∼100
∼100
∼50
∼300
∼50
∼100
∼100
∼100
∼50
23
CH3
H
b
2g
3c
3d
3e
3f
H
∼100
12
CH3
−CH2−
H
CH3
0.87
1.7
2.6
33
4.4
16
CH3
H
10
CH3
H
1.7
6.8
3g
H
binding Ki (μM)
functional IC50 (μM)
R″
α4β2
α4β4
α3β4
α7/5-HT3A
α4β2
α3β4
9a
9b
9c
9d
9e
9f
ethyl
∼300
∼30
∼300
∼100
∼300
>300
>300
>300
>300
>300
>300
>300
n-butyl
cyclohexyl
phenyl
3-methoxyphenyl
3,4-dimethoxy-
phenyl
a
The Ki values for the compounds at the rat α4β2, α4β4, and α3β4 nAChR subtypes were obtained in a [3H]epibatidine binding assay and the Ki
values for the compounds at the α7/5-HT3A chimera were determined in a [3H]MLA binding assay. The functional IC50 values for the compounds at
the mouse α4β2 nAChR and the rat α3β4 nAChR were obtained in the FLIPR Membrane Potential Blue assay. (S)-Nicotine (EC70−EC80 for the
respective receptors, i.e., 1 μM for α4β2 and 3 μM for α3β4) was used as agonist in the functional experiments. The data were the means of 3−5
individual experiments performed in duplicate. The complete pharmacological data (including SEM values) are given in the Supporting Information.
b
Because a complete concentration−inhibition curve could not be obtained for the compound in the concentration ranges tested, the IC50 value
given is an estimate.
9676
dx.doi.org/10.1021/jm4013592 | J. Med. Chem. 2013, 56, 9673−9682

Journal of Medicinal Chemistry
Article
Figure 3. Correlation between binding and functional data for compound series 1−3 at the α4β2 nAChR. In (A), data from all three series are
shown together, whereas (B−D) show them individually.
on mouse α4β2 nAChR and rat α3β4 signaling in the FMP
Blue assay (Table 1). The overall correlation between binding
affinities and functional inhibitory potencies of the 15
compounds at the α4β2 nAChR was not impressive (Figure
3A). Interestingly, considerably more pronounced correlations
were observed for each of the three compound classes
separately, in particular when the data for 2e and 3c (which
constituted outliers in the 2c−g and 3c−g series) were omitted
from the comparisons, giving R2 values of 0.99 and 0.97,
respectively (Figure 3B,C,D).
In terms of structure−activity relationships, compounds with
all four rings intact (1c−g) retain some affinity for α7, but this
is eliminated as the size of the ligands is reduced, as exemplified
by the direct analogues of erysodine (ligands 1e−3e): With all
four rings intact, the affinity at α7 of compound 1e of 20 μM is
in the same range as the 7.5 μM reported for erysodine.6 In
contrast, the binding affinity for erysodine at α4β2 drops from
50 nM to 8.5 μM for 1e (Table 1). Deleting the A-ring gives 2e
with negligible affinity for α7 and α4β2 (∼500 and ∼100 μM,
respectively). Removal of the B-ring in 1e (3e) restores affinity
for α4β2 to the same level as for 1e.
In general, no clear picture appeared with respect to the
effect of the substitution pattern on the catechol unit when it
comes to affinity for α4β2. Introduction of a single methoxy
group in the R′ position yields some of the least active
analogues in all three compound series (1e−3e), most
pronounced in the virtually inactive 2e analogue. Generally,
the dioxolane unit performed well for all three compounds in
the series (1d−3d). Compound 2d has the highest affinity for
α4β2 of all compounds in this study (0.55 μM), but this
compound also exhibits affinity for α4β4.
α4β2 is in the nM range with greater than 300-fold selectivity
toward the other subtypes tested in this study as well as a
virtually inactive analogue (1c). Compound 1c has the lowest
affinity for all four subtypes of all compounds with the A- and
B-rings intact, but as the A ring is removed to give 2c, affinity
for α4β2 increases to 5.5 μM and deletion of the B-ring gives
3c, which is quite potent and the most selective compound in
the series. In comparison, the parent alkaloid erysotrine has
been reported to have a binding affinity at α4β2 of 0.6 μM.6
Thus, 3c is an interesting new lead compound with low
molecular weight that allows for further optimization.
To probe the potential in 3c, the 1-position of the
tetrahydroisoquinoline ring system where the A- and B-rings
are attached in the Erythrina alkaloids was modified by
introducing different substituents: three ligands with an alkyl
side chain (9a−c) and three other derivatives bearing an
aromatic moiety (9d−f), see Scheme 5. Unfortunately,
substitution at this position on 3c was detrimental to activity
at α4β2 as shown in Table 1. With the exception of 9b and 9d,
the ligands were virtually devoid of affinity for α4β2. This is
somewhat perplexing but could be taken as an indication of a
different binding mode of 3c to the nAChR than that of the
parent alkaloid.
A common feature of all ligands tested in this study was the
deletion of the methoxy group on the A ring. This decision was
supported by a previously published computational and
mutational study on the binding mode of the erythrinanes.6
Therein it was concluded that the tertiary nitrogen and the
catecol unit are key interaction points for the erythrinanes: both
of these structural elements are retained in the present
compound series. As mentioned above, no clear picture
appeared with respect to the substitution pattern on the
catechol unit when it comes to affinity for α4β2. This can be
The three dimethoxy analogues comprise one of the most
potent antagonists in the entire series (3c) where affinity for
9677
dx.doi.org/10.1021/jm4013592 | J. Med. Chem. 2013, 56, 9673−9682

Journal of Medicinal Chemistry
Article
interpreted in a number of different ways. Assuming that the
tetracyclic derivatives 1c−g adopts a similar conformation as
the natural products, the deletion of the double bonds and
removal of the methoxy group on the A-ring should not be
detrimental to the affinity for α4β2, but we see a dramatic
decrease. Thus, these structural motifs clearly play an important
role in the interaction of erythrinanes with the nAChRs. The
three and two ring analogues (2c−g and 3c−g) may adopt a
different mode of binding due to the smaller size of the ligands,
thereby making a direct comparison with the parent natural
products difficult.
Furthermore, we recently obtained an X-ray structure of the
acetylcholine binding protein (AChBP) in complex with
DHβE, with a different binding mode.21 This structure suggests
that the methoxy group on the A-ring of DHβE plays a crucial
role due to its interaction with a tightly bound water molecule
in the protein. With the absence of the methoxy substituent on
the A-ring, this interaction is not possible and may account for
the reduced affinity of the investigated compounds relative to
the parent natural products. It is therefore likely that the
compounds presented here adopt a binding mode different to
that observed for DHβE in the X-ray structure, and this would
explain the lack of affinity of the series 9 compounds.
have Ki values of 47 and 130 μM on SERT and NET,
respectively (see Figure 4). Surprisingly, no inhibition of DAT
could be detected (Ki > 10 mM). Thus, the inhibitory potency
of 3c is much lower at SERT and NET compared to
prototypical antidepressant drugs such as the dual acting
SERT/NET reuptake inhibitors duloxetine and venlafaxine,
which have Ki values for SERT and NET in the nM range.30
Tetrahydroisoquinoline 3c has been known in the chemical
literature for many years and was also isolated from natural
sources in 1986 and named O-methylcorypalline.31 It is
reported to be a weak inhibitor of monoamine oxidase A and
B: Ki = 27 and 29 μM, respectively.32 Because inhibitors of
monoamine oxidase have proven efficacy in the treatment of
depression,33 3c has an interesting pharmacological profile for
antidepressant activity.
Thus, we decided to investigate the antidepressant-like effect
of 3c in vivo using the mouse forced swim test.34 In this
behavioral test, a mouse is placed in an inescapable tank filled
with water and the distance traveled by the animal is measured.
An increase in swimming distance is interpreted as an
antidepressant effect of the tested compound. We examined
the effect of 3c at 10 and 30 mg/kg and found that it exhibited
a significant antidepressant-like effect at 30 mg/kg, see Figure 5.
A link between depression and nicotinic cholinergic
neurotransmission is evident from numerous clinical and
preclinical studies.22 Antidepressant effects in humans and
antidepressant-like effects in rodents have been reported with
both agonism and antagonism of nAChRs.23 Similarly,
enhanced antidepressant-like action of conventional antide-
pressants in rodents has been shown with both nAChR
agonists24,25 and antagonists.26,27 Given the antidepressant
effect of the known nAChR antagonist DHβE in vivo,26,28 we
were intrigued by the possibility of 3c having similar
antidepressant activity.
An interesting feature of compound 3c is that the structural
motive of dopamine is imbedded in the C and D rings, see
Figure 4. This prompted us to investigate the activity of 3c at
Figure 5. Results from mouse forced swim test of 3c.
The relatively high dose needed to evoke the antidepressant-
like effect raises a question of whether this is mediated solely via
nAChRs or via a combination of one or more of the different
targets mentioned above. The 30 mg/kg dose showing
behavioral activity is equivalent to approximately 10-fold
lower DHβE doses (3 mg/kg). Indeed, this dose of DHβE
has been demonstrated several times to produce antidepres-
sant-like effects in mice.28 Thus, we speculate that the effect is
at least in part mediated via the nAChRs, but this will have to
be investigated further before any definite conclusions can be
made on the mechanism of action.
Figure 4. (A) Structure of 3c and dopamine showing that 3c is a
restricted analogue of dopamine. (B) Dose−response curves from
representative experiments of inhibition by 3c of [3H]neurotransmitter
uptake in COS7 cells expressing human SERT, NET, or DAT. Data
points represent the mean SEM from triplicate determinations. The
inhibitory potency of 3c was determined from 4−6 independent
CONCLUSIONS
■
The present study describes the design, synthesis, and
biological evaluation of analogues of the tetracyclic Erythrina
alkaloids as nAChR antagonists. Reducing the complexity of the
natural products via the sequential deletion of rings gave an
initial series of 15 ligands. Compound 3c (a previously reported
natural product) emerged as a subtype selective competitive
antagonist of the α4β2 nAChR with antidepressant effect in
vivo. Initial attempts to investigate the potential this scaffold via
homologation at the 1-position of the tetrahydroisoquinoline
ring systems were detrimental to the affinity toward α4β2, but
3c remains an interesting and readily manipulated lead
structure.
experiments, each performed in triplicate (mean
SEM; in μM):
hSERT 46.7 7.1, hNET, 128.9 15.0 hDAT >10000).
some classical targets for antidepressants, the plasma membrane
transporters for the monoamine neurotransmitters serotonin,
norepinephrine, and dopamine (SERT, NET, and DAT,
respectively).29 The inhibitory potency of ligand 3c was
determined at SERT, NET, and DAT in in vitro [3H]-
neurotransmitter uptake inhibition assays, and we found 3c to
9678
dx.doi.org/10.1021/jm4013592 | J. Med. Chem. 2013, 56, 9673−9682

Journal of Medicinal Chemistry
Article
stirred at this temperature for 30 min. The solution was made alkaline
by addition of an aqueous solution of NaOH (2 M), and the aqueous
layer was extracted with EtOAc. The organic layer was washed with
brine, dried over Na2SO4, filtered, and concentrated in vacuo. The
crude material was purified by column chromatography on silica gel to
provide the pure corresponding cis-erythrinane 1.
EXPERIMENTAL SECTION
■
1.1. Medicinal Chemistry: Material and Methods. Starting
materials and reagents were obtained from commercial suppliers and
used without further purifications. Syringes which were used to
transfer anhydrous solvents or reagents were purged with nitrogen
prior to use. THF was continuously refluxed and freshly distilled from
sodium/benzophenone under nitrogen. Other solvents were analytical
or HPLC grade and were used as received. Yields refer to isolated
Typical Procedure 4: Synthesis of cis-Erythrinanes 1b and 1c. To
a solution of the appropriate phenethylamine 4 (1 equiv) in toluene
(10−35 mL) was added ethyl 2-cyclohexanoneacetate (1.05 equiv) at
room temperature. The mixture was heated under reflux conditions for
12 h. After cooling to room temperature, an aqueous solution of HI
(7.6 M) was added dropwise then the reaction mixture was stirred for
2 h. After addition of an aqueous solution of NaHSO3 (2 M), the
mixture was extracted twice with EtOAc and the combined organic
layers were washed with brine, dried over Na2SO4, filtered, and
concentrated in vacuo. The resulting crude was then dissolved in dry
THF (5 mL) and slowly added to a suspension of LiAlH4 (2.5 equiv)
in dry THF (10 mL) at 0 °C. The mixture was allowed to warm up to
room temperature and stirred at this temperature for 12 h then cooled
to 0 °C. An aqueous solution of Rochelle salt (2 M) was carefully
added and the reaction mixture stirred at this temperature for 30 min.
The solution was made alkaline by addition of an aqueous solution of
NaOH (2 M), and the aqueous layer was extracted with EtOAc. The
organic layer was washed with brine, dried over Na2SO4, filtered, and
concentrated in vacuo. The crude material was purified by column
chromatography on silica gel to provide the pure corresponding cis-
erythrinane 1.
1
compounds estimated to be >95% pure as determined by H NMR
and LC-MS. Thin-layer chromatography (TLC) was carried out on
silica gel 60 F254 plates from Merck (Germany). Visualization was
accomplished by UV lamp (254 nm) or with either ninhydrine dip
with heat or iodine on silica as an indicator. Flash column
chromatography was performed on chromatography grade, silica 60
Å particle size 35−70 μm from Fisher Scientific using the solvent
1
system as stated. H NMR and 13C NMR spectra were recorded on
Varian 300 (Mercury and Gemini) instruments, using CDCl3 or
DMSO-d6 as solvents and TMS as internal standard. Coupling
constants (J values) are given in hertz (Hz). Multiplicities of 1H NMR
signals are reported as follows: s, singlet; d, doublet; dd, doublet of
doublets; dt, doublet of triplets; t, triplet; q, quartet; sept, septet;
multiplet; br, broad signal. Melting points (mp) were measured using a
MPA100 Optimelt melting point apparatus and are uncorrected.
Elemental analyses were performed at the Department of Physical
Chemistry, University of Vienna, Austria. High-resolution mass spectra
(HRMS) were obtained using a Micromass Q-TOF 2 instrument. The
following abbreviations are used: MeOH, methanol; DCM, dichloro-
methane; THF, tetrahydrofuran; EtOAc, ethyl acetate; AcOH, acetic
acid; TEA, triethylamine.
1.2. Typical Procedures and Analytical Data of Representa-
tive Synthesized Compounds. Typical Procedure 1: Synthesis of
Tetrahydroisoquinolines 3. To a solution of phenethylamine 4 (1
equiv) in formic acid (7−13 mL) was added a 37% aqueous solution of
formaldehyde (5 equiv) at room temperature. The mixture was heated
at 100 °C for 2 h then cooled to 0 °C and made alkaline by addition of
an aqueous solution of NaOH (2 M). The aqueous layer was extracted
twice with EtOAc, and the combined organic layers were washed with
brine, dried over Na2SO4, filtered, and concentrated in vacuo. The
crude material was purified by column chromatography on silica gel to
provide pure tetrahydroisoquinoline 3.
Typical Procedure 2: Synthesis of Indolizidines 2. To a solution of
amide 8 (1 equiv) in toluene (20−150 mL) was added POCl3 (10
equiv) at room temperature. The mixture was heated under reflux
conditions for 4.5 h, then solvents were removed in vacuo. To the
crude material dissolved in a mixture EtOH−AcOH (10:1, 11−66
mL) was added NaBH4 (2 equiv) portionwise at 0 °C. The mixture
was allowed to warm up to room temperature and stirred at this
temperature for 16 h. After addition of water, volatiles were removed
and the aqueous layer was extracted three times with DCM. The
combined organic layers were successively washed with an aqueous
solution of NaOH (2 M) and brine, dried over Na2SO4, filtered, and
concentrated in vacuo. The resulting crude was purified by column
chromatography on silica gel to provide pure indolizidine 2.
Typical Procedure 3: Synthesis of cis-Erythrinanes 1a and 1d. To
a solution of the appropriate phenethylamine 4 (1 equiv) in
acetonitrile (3−4 mL) were added successively AlMe3 (2 M in
toluene, 2 equiv) dropwise, Sc(OTf)3 (12.5 mol %), and ethyl 2-
cyclohexanoneacetate (1.05 equiv) at 0 °C. The mixture was allowed
to warm up to room temperature and stirred at this temperature for 17
h. After cooling to 0 °C, TfOH (3.5 equiv) was added dropwise, then
the reaction mixture was allowed to warm up to room temperature and
stirred at this temperature for 2 h. After addition of a saturated
aqueous solution of NaHCO3 at 0 °C, the mixture was extracted twice
with EtOAc and the combined organic layers were washed with water,
dried over Na2SO4, filtered, and concentrated in vacuo. The resulting
crude was then dissolved in dry THF (5 mL) and slowly added to a
suspension of LiAlH4 (2.5 equiv) in dry THF (10 mL) at 0 °C. The
mixture was allowed to warm up to room temperature and stirred at
this temperature for 12 h then cooled to 0 °C. An aqueous solution of
Rochelle salt (2 M) was carefully added and the reaction mixture
6,7-Dimethoxy-2-methyl-1,2,3,4-tetrahydroisoquinoline (3c).18a
Starting from phenethylamine 4e (1.00 g, 5.5 mmol), following
typical procedure 1 and using EtOAc−MeOH−TEA (40:10:1) as
eluent, tetrahydroisoquinoline 3c was obtained as a pale-yellow solid
(1.13 g, 99%); mp 69−71 °C; Rf = 0.25 (EtOAc−MeOH−TEA,
1
40:10:1). H NMR (300 MHz, CDCl3): δ 6.56 (s, 1H), 6.47 (s, 1H),
3.81 (s, 3H), 3.80 (s, 3H), 3.47 (br s, 2H), 2.81 (t, J = 5.8, 2H), 2.63
(t, J = 5.8, 2H), 2.42 (s, 3H). 13C NMR (75 MHz, CDCl3): δ 147.5,
147.2, 126.7, 125.8, 111.4, 109.4, 57.9, 56.2, 56.1, 53.3, 46.4, 29.2. Anal.
Calcd For C12H17NO2: C, 69.54; H, 8.27; N, 6.76. Found: C, 69.54; H,
8.24; N, 6.41.
6,7-Methylenedioxy-2-methyl-1,2,3,4-tetrahydroisoquinoline
(3d).35 Starting from phenethylamine 4f (0.50 g, 3.0 mmol), following
typical procedure 1 and using EtOAc−MeOH−TEA (40:10:1) as
eluent, tetrahydroisoquinoline 3d was obtained as an off-white solid
(0.42 g, 72%); mp 53−55 °C; Rf = 0.35 (EtOAc−MeOH−TEA,
1
40:10:1). H NMR (300 MHz, CDCl3): δ 6.52 (s, 1H), 6.44 (s, 1H),
5.83 (s, 2H), 3.43 (s, 2H), 2.79 (t, J = 5.8, 2H), 2.60 (t, J = 5.8, 2H),
2.40 (s, 3H). 13C NMR (75 MHz, CDCl3): δ 145.8, 145.5, 127.4,
126.6, 108.3, 106.2, 100.5, 58.0, 52.9, 46.1, 29.3. Anal. Calcd For
C11H13NO2: C, 69.09; H, 6.85; N, 7.32. Found: C, 68.89; H, 6.55; N,
7.09.
8,9-Dimethoxy-1,2,3,5,6,10b-hexahydropyrrolo[2,1-a]-
isoquinoline (2c).17 Starting from amide 8c (1.52 g, 5.7 mmol),
following typical procedure 2 and using a gradient elution (DCM−
MeOH, 6:1 to 4:1), indolizidine 2c was obtained as a off-white solid
1
(0.68 g, 52%); mp 87−89 °C; Rf = 0.45 (DCM−MeOH, 4:1). H
NMR (300 MHz, CDCl3): δ 6.60 (s, 1H), 6.56 (s, 1H), 3.85 (s, 6H),
3.52 (br t, J = 8.2, 1H), 3.14−3.23 (m, 1H), 2.95−3.12 (m, 2H), 2.61−
2.81 (m, 3H), 2.29−2.41 (m, 1H), 1.67−2.03 (m, 3H). 13C NMR (75
MHz, CDCl3): δ 147.5, 147.4, 130.5, 126.1, 111.4, 108.9, 63.1, 56.3,
56.2, 53.5, 48.6, 31.0, 28.2, 22.6; Anal. Calcd For C14H19NO2: C,
72.07; H, 8.21; N, 6.00. Found: C, 72.41; H, 7.94; N, 5.80.
8,9-Methylenedioxy-1,2,3,5,6,10b-hexahydropyrrolo[2,1-a]-
isoquinoline (2d). Starting from amide 8d (0.64 g, 2.4 mmol),
following typical procedure 2 and using DCM−MeOH (3:1) as eluent,
indolizidine 2d was obtained as a pale-yellowish oil (0.37 g, 69%)
which solidified when stored in the refrigerator; mp 62−64 °C; Rf =
1
0.4 (DCM−MeOH, 10:3). H NMR (300 MHz, CDCl3): δ 6.57 (s,
1H), 6.54 (s, 1H), 5.88 (s, 2H), 3.33 (br t, J = 8.8, 1H), 2.96−3.20 (m,
3H), 2.73 (br dt, J = 16.2 and J = 3.6, 1H), 2.47−2.64 (m, 2H), 2.23−
2.35 (m, 1H), 1.62−2.02 (m, 3H). 13C NMR (75 MHz, CDCl3): δ
145.9, 145.8, 132.1, 127.4, 108.6, 106.1, 100.8, 63.7, 53.6, 48.8, 30.8,
9679
dx.doi.org/10.1021/jm4013592 | J. Med. Chem. 2013, 56, 9673−9682

Journal of Medicinal Chemistry
Article
29.1, 22.4. HRMS (m/z): [M + H]+ calcd for [C13H16NO2]+
218.1181; found 218.1179.
Ci/mmol; SERT assays) or [3H]-dopamine (48 Ci/mmol; NET and
DAT assays) were added, giving a final substrate concentration of 50
nM. Uptake was allowed to proceed for 10 min and terminated by
aspiration and rapid washing of wells three times with PBSCM.
Amount of accumulated radioligand was determined by solubilizing
cells in scintillant (MicroScint20), followed by counting of plates in a
Packard TopCounter. Nonspecific uptake was determined in parallel
by measuring uptake in nontransfected cells. IC50 values were
converted to Ki values using the Cheng−Prusoff equation.
Mouse Forced Swim Test. Method: Mice (n = 7−8) were
individually placed in a beaker (16 cm in diameter) filled with 20
cm of water maintained at 23.5−24.5 °C. Total swim distance during
the 6 min test period was automatically recorded by a camera mounted
above the cylinders and stored on a computer equipped with
Ethovision (Noldus, Holland). 3c was dissolved in saline (0.9%
NaCl) and given subcutaneously 20 min prior to testing in an injection
volume of 10 mL/kg. Data analysis: The first minute was omitted from
the data before statistical analysis. This is because animals generally
swim extensively for the first minute, irrespective of treatment; hence,
any true treatment effect only becomes apparent after one minute.
Swim distance was analyzed using a one-way analysis of variance
(ANOVA) and followed by pairwise comparisons of the predicted
means using the Planned Comparisons procedure. To ensure variance
homogeneity and normality, a requirement of the ANOVA approach,
data were log-transformed before statistical analysis. Differences were
considered significant for p < 0.05. Results: The ANOVA revealed a
significant main effect of 3c treatment (F2,20 = 10.47; p < 0.001).
Planned Comparisons showed that swim distance was significantly
increased by 30 mg/kg 3c (p < 0.001).
16-Isopropoxy-15-methoxy-cis-erythrinane (1a). Starting from
phenethylamine 4c (0.35 g, 1.7 mmol), following typical procedure
3 and using DCM−MeOH−TEA (100:15:1) as eluent, cis-erythrinane
1a was obtained as a pale-yellow oil (0.1 g, 20%); Rf = 0.55 (DCM−
1
MeOH−TEA, 50:5:1). H NMR (300 MHz, CDCl3): δ 6.67 (s, 1H),
6.49 (s, 1H), 4.45 (sept, J = 6.0, 1H), 3.82 (s, 3H), 2.97−3.23 (m,
4H), 2.84 (dt, J = 10.5 and J = 3.0, 1H), 2.19−2.33 (m, 2H), 1.87−
1.98 (m, 1H), 1.34 (d, J = 6.1, 6H), 1.19−1.79 (m, 9H). 13C NMR (75
MHz, CDCl3): δ 148.5, 145.5, 136.1, 126.9, 115.2, 109.9, 71.3, 64.7,
56.5, 46.6, 43.9, 40.8, 36.0, 29.2, 28.9, 25.3, 22.51, 22.47, 21.7, 21.6.
HRMS (m/z): [M + H]+ calcd for [C20H30NO2]+ 316.2277; found
316.2273.
15-Isopropoxy-16-methoxy-cis-erythrinane (1b). Starting from
phenethylamine 4d (0.35 g, 1.7 mmol), following typical procedure
4 and using DCM−MeOH−TEA (100:15:1) as eluent, cis-erythrinane
1b was obtained as a pale-yellow oil (0.2 g, 38%); Rf = 0.55 (DCM−
1
MeOH−TEA, 50:5:1). H NMR (300 MHz, CDCl3): δ 6.74 (s, 1H),
6.50 (s, 1H), 4.46 (sept, J = 6.1, 1H), 3.81 (s, 3H), 3.01−3.24 (m,
4H), 2.85 (dt, J = 10.4 and J = 3.0, 1H), 2.16−2.36 (m, 2H), 1.88−
1.98 (m, 1H), 1.35 (dd, J = 6.1 and J = 2.2, 6H), 1.21−1.80 (m, 9H).
13C NMR (75 MHz, CDCl3): δ 148.8, 145.3, 136.2, 128.0, 115.1,
112.0, 72.2, 64.6, 56.1, 46.6, 44.0, 40.8, 36.1, 29.3, 29.0, 25.4, 22.54,
22.51, 21.9, 21.8.HRMS (m/z): [M + H]+ calcd for [C20H30NO2]+
316.2277; found 316.2261.
15,16-Dimethoxy-cis-erythrinane (1c).36 Starting from phenethyl-
amine 4e (1.36 g, 7.5 mmol), following typical procedure 4 and using a
gradient elution (DCM−MeOH, 300:15 to DCM−MeOH−TEA,
300:15:1), cis-erythrinane 1c was obtained as a pale-orange oil (0.74 g,
34%); Rf = 0.3 (DCM−MeOH−TEA, 300:15:1). 1H NMR (300
MHz, CDCl3): δ 6.70 (s, 1H), 6.51 (s, 1H), 3.88 (s, 3H), 3.85 (s, 3H),
3.03−3.22 (m, 4H), 2.87 (dt, J = 10.4 and J = 2.7, 1H), 2.19−2.33 (m,
2H), 1.89−2.03 (m, 1H), 1.23−1.82 (m, 9H). 13C NMR (75 MHz,
CDCl3): δ 147.3, 147.1, 135.9, 126.9, 111.4, 110.0, 62.2, 56.4, 56.0,
46.6, 44.0, 40.7, 36.1, 29.4, 29.0, 25.5, 21.8, 21.7. HRMS (m/z): [M +
H]+ calcd for [C18H26NO2]+ 288.1964; found 288.1961.
ASSOCIATED CONTENT
* Supporting Information
■
S
All general procedures and characterization for all synthesized
compounds, copies of 1H and 13C NMR spectra for compounds
1a−g, 2a−g, 3a−g, 7a−f, 8a−d, 9a−f, and 10, as well as tables
containing the full data set from the nAChR-pharmacology.
This material is available free of charge via the Internet at
http://pubs.acs.org.
15,16-Methylenedioxy-cis-erythrinane (1d).13c Starting from
phenethylamine 4f (0.25 g, 1.5 mmol), following typical procedure 3
and using DCM−MeOH−TEA (100:15:1) as eluent, cis-erythrinane
1d was obtained as a colorless oil (0.23 g, 56%); Rf = 0.4 (DCM−
AUTHOR INFORMATION
Corresponding Author
*Phone: (+45) 35336487. E-mail: jesper.kristensen@sund.ku.
■
1
MeOH−TEA, 100:15:1). H NMR (300 MHz, CDCl3): δ 6.71 (s,
1H), 6.48 (s, 1H), 5.87 (d, J = 1.4, 1H), 5.86 (d, J = 1.4, 1H), 2.98−
3.23 (m, 4H), 2.84 (dt, J = 10.5 and J = 3.0, 1H), 2.15−2.35 (m, 2H),
1.86−1.97 (m, 1H), 1.22−1.80 (m, 9H). 13C NMR (75 MHz, CDCl3):
δ 145.9, 145.4, 137.4, 128.0, 108.4, 106.0, 100.6, 65.1, 46.5, 44.0, 40.6,
36.1, 29.3, 28.8, 25.4, 22.3, 21.8. HRMS (m/z): [M + H]+ calcd for
[C17H22NO2]+ 272.1651; found 272.1649.
dk.
Present Address
§Medicinal Chemistry Research, H. Lundbeck A/S, Ottiliavej 9,
2500 Valby, Denmark.
1.3. In Vitro and in Vivo Pharmacology. nAChR Binding
Assays. The binding properties of the compounds at rat α4β2, α4β4,
and α3β4 nAChRs were determined in a [3H]epibatidine binding assay
and at the α7/5-HT3A chimera in a [3H]MLA binding assay as
previously described.19 IC50 values were converted to Ki values using
the Cheng−Prusoff equation.
The FLIPR Membrane Potential Assay. The functional properties
of the compounds were determined at the mouse α4β2 and rat α3β4
nAChR subtypes essentially as previously described.37
SERT, NET, and DAT Uptake Inhibition Assay. COS7 cells
(American Type Culture Collection, Manassas, VA) were cultured
and transiently transfected with mammalian expression plasmids
containing human SERT (pcDNA3.1-hSERT), human NET (pCI-
IRES-neo-hNET), or human DAT (pCI-IRES-neo-hDAT) as detailed
previously.38 Uptake inhibition assays were performed 40−48 h after
transfection in white 96-well tissue culture plates essentially as
described previously.38 Briefly, cells were preincubated in phosphate-
buffered saline (PBS; 137 mM NaCl, 2.7 mM KCl, 4.3 mM Na2HPO4,
1.4 mM KH2PO4, pH 7.4) containing 0.1 mM CaCl2 and 0.5 mM
MgCl2 (PBSCM) with 0 or increasing concentrations of 3c. In all NET
and DAT assays, the PBSCM buffer was supplemented with 1 mM
ascorbic acid. After 30 min, PBSCM containing [3H]-serotonin (28
Author Contributions
∥F.C. and A.A.J. contributed equally
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the Danish Council for Independent Research−
Medical Sciences, the Lundbeck Foundation, and the Novo
Nordisk Foundation for financial support.
REFERENCES
■
(1) (a) Jensen, A. A.; Frolund, B.; Liljefors, T.; Krogsgaard-Larsen, P.
Neuronal Nicotinic Acetylcholine Receptors: Structural Revelations,
Target Identifications, and Therapeutic Inspirations. J. Med. Chem.
2005, 48, 4705−4745. (b) Arneric, S. P.; Holladay, M.; Williams, M.
Neuronal Nicotinic Receptors: A Perspective on Two Decades of
Drug Discovery Research. Biol. Pharmacol. 2007, 74, 1092−1101.
(c) Steinlein, O. K. Ion Channel Mutations in Neuronal Diseases: A
Genetics Perspective. Chem. Rev. 2012, 112, 6334−6352. (d) Lemoine,
9680
dx.doi.org/10.1021/jm4013592 | J. Med. Chem. 2013, 56, 9673−9682

Journal of Medicinal Chemistry
Article
D.; Jiang, R.; Taly, A.; Chataigneau, T.; Specht, A.; Grutter, T. Ligand-
Gated Ion Channels: New Insights into Neurological Disorders and
Ligand Recognition. Chem. Rev. 2012, 112, 6285−6318.
(11) Cabedo, N.; Protais, P.; Cassels, B. K.; Cortes, D. Synthesis and
Dopamine Receptor Selectivity of the Benzyltetrahydroisoquinoline, I-
(+)-nor-Roefractine. J. Nat. Prod. 1998, 61, 709−712.
(12) For a recent review on the synthesis of Erythrina-type
compounds, see: Riemann, E. Synthesis Pathways to Erythrina
Alkaloids and Erythrina Type Compounds. Prog. Chem. Org. Nat.
Prod. 2007, 88, 1−62.
(2) (a) Coe, J. W.; Brooks, P. R.; Vetelino, M. G.; Wirtz, M. C.;
Arnold, E. P.; Huang, J.; Sands, S. B.; Davis, T. I.; Lebel, L. A.; Fox, C.
B.; Shrikhande, A.; Heym, J. H.; Schaeffer, E.; Rollema, H.; Lu, Y.;
Mansbach, R. S.; Chambers, L. K.; Rovetti, C. C.; Schulz, D. W.;
Tingley, F. D., III; O’Neill, B. T. Varenicline: An α4β2 Nicotinic
Receptor Partial Agonist for Smoking Cessation. J. Med. Chem. 2005,
48, 3474−3477. (b) Bertrand, D.; Bertrand, S.; Cassar, S.; Gubbins, E.;
Li, J.; Gopalakrishnan, M. Positive Allosteric Modulation of the α7
Nicotinic Acetylcholine Receptor: Ligand Interactions with Distinct
Binding Sites and Evidence of a Prominent Role of the M2−M3
Segment. Mol. Pharmacol. 2008, 74, 1407−1416.
(13) For recent examples on the stereocontrolled syntheses of
Erythrina-type products, see: (a) Zhang, F.; Simpkins, N. S.; Blake, A.
J. New Approaches for the Synthesis of Erythrinan Alkaloids. Org.
Biomol. Chem. 2009, 7, 1963−1979. (b) He, Y.; Funk, R. L. Total
Syntheses of ( )-β-Erythroidine and ( )-8-oxo-β-Erythroidine by an
Intramolecular Diels−Alder Cycloaddition of a 2-Amidoacrolein. Org.
Lett. 2006, 8, 3689−3692. (c) Reimann, E.; Ettmayr, C. An Improved
Stereocontrolled Route to cis-Erythrinanes by Combined Intra-
molecular Strecker and Bruylants Reaction. Monatsh. Chem. 2004,
135, 1143−1155.
(3) (a) Dwoskin, L. P.; Crooks, P. A. Competitive Neuronal
Nicotinic Receptor Antagonists: A New Direction for Drug Discovery.
J. Pharmacol. Exp. Ther. 2001, 298, 395−402. (b) Iturriaga-Vas
́
quez,
P.; Perez, E. G.; Slater, E. Y.; Bermudez, I.; Cassels, B. K. Aporphine
́
(14) Tietze, L. F.; Tolle, N.; Noll, C. Highly Efficient Domino
̈
Metho Salts as Neuronal Nicotinic Acetylcholine Receptor Blockers.
Bioorg. Med. Chem. 2007, 15, 3368−3372. (c) For recent studies on
antagonists of α7 nAChR, see: Ambrus, J. I.; Halliday, J. I.; Kanizaj, N.;
Absalom, N.; Harpsøe, K.; Balle, T.; Chebib, M.; McLeod, M. D.
Covalent Attachment of Antagonists to the α7 Nicotinic Acetylcholine
Receptor: Synthesis and Reactivity of Substituted Maleimides. Chem.
Commun. 2012, 6699−6701. (d) For recent studies on selective
antagonists of α4β2 nAChR, see: Faundez-Parraquez, M.; Farias-
Rabelo, N.; Gonzalez-Gutierrez, J. P.; Etcheverry-Berrios, A.; Alzate-
Reaction for the Synthesis of the Erythrina and B-Homoerythrina
Alkaloid Skeleton. Synlett 2008, 525−528.
(15) Isobe, K.; Mohri, K.; Tokoro, K.; Fukushima, C.; Higuchi, F.;
Taga, J.-I.; Tsuda, Y. Selective Cleavage of Aromatic Rings by
Ozonolysis. I. Application to o-Dimethoxybenzene Derivatives. Chem.
Pharm. Bull. 1988, 36, 1275−1282.
̌
(16) Smidrkal, J. Synthesis of Fagaronine. Collect. Czech. Chem.
Commun. 1988, 53, 3184−3192.
(17) Bailey, K. R.; Ellis, A. J.; Reiss, R.; Snape, T. J.; Turner, N. J.
Template-Based Mnemonic for Monoamine Oxidase (MAO-N)
Catalyzed Reactions and its Application Towards the Chemo-
Enzymatic Deracemisation of the Alkaloid ( )-Crispine A. Chem.
Commun. 2007, 3640−3642.
Morales, J.; Adasme-Carreno, F.; Varas, R.; Bermudez, I.; Iturriaga-
̃
́
Vasquez, P. Neonicotinic Analogues: Selective Antagonists for α4β2
Nicotinic Acetylcholine Receptors. Bioorg. Med. Chem. 2013, 21,
2687−2694.
(18) (a) Ruchirawat, S.; Chaisupakitsin, M.; Patranuwatana, N.;
Cashaw, J. L.; Davis, V. E. A Convenient Synthesis of Simple
Tetrahydroisoquinolines. Synth. Commun. 1984, 14, 1221−1228. (b)
(4) (a) Boekelheide, V.The Erythrina Alkaloids. In The Alkaloids:
Chemistry and Physiology; Academic Press: New York, 1960; Vol. 7, pp
201−227. (b) Reimann, E. Synthesis Pathways to Erythrina Alkaloids
and Erythrina Type Compounds. In Progress in the Chemistry of Organic
Natural Products; Springer: New York, 2007; Vol. 88, pp 1−62.
(5) Erysovine and Erythraline have not yet been characterized on the
nAChRs.
For a recent review about the Pictet−Spengler reaction, see: Stockigt,
̈
J.; Antonchick, A. P.; Wu, F.; Waldmann, H. The Pictet−Spengler
Reaction in Nature and in Organic Chemistry. Angew. Chem., Int. Ed.
2011, 50, 8538−8564.
(6) Iturriaga-Vas
́
quez, P.; Carbone, A.; García-Beltran
́
, O.;
(19) Jensen, A. A.; Mikkelsen, I.; Frølund, B.; Bauner-Osborne, H.;
̈
Livingstone, P. D.; Biggin, P. C.; Cassels, B. K.; Wonnacott, S.;
Zapata-Torres, G.; Bermudez, I. Molecular Determinants for
Competitive Inhibition of α4β2 Nicotinic Acetylcholine Receptors.
Mol. Pharmacol. 2010, 78, 366−375.
Falch, E.; Krogsgaard-Larsen, P. Carbamoylcholine Homologs: Novel
and Potent Agonists at Neuronal Nicotinic Acetylcholine Receptors.
Mol. Pharmacol. 2003, 64, 865−875.
(20) Armishaw, C. J.; Singh, N.; Medina-Franco, J. L.; Clark, R. J.;
Scott, K. C. M.; Houghten, R. A.; Jensen, A. A. A Synthetic
Combinatorial Strategy for Developing α-Conotoxin Analogs as
Potent α7 Nicotinic Acetylcholine Receptor Antagonists. J. Biol.
Chem. 2010, 285, 1809−1821.
(21) Shahsavar, A.; Kastrup, J. S.; Nielsen, E. Ø.; Kristensen, J. L.;
Gajhede, M.; Balle, T. Crystal Structure of Lymnaea stagnalis AChBP
Complexed with the Potent nAChR Antagonist DHßE Suggests a
Unique Mode of Antagonism. PLoS One 2012, 7, e40757.
(22) Quattrocki, E.; Baird, A.; Yurgelun-Todd, D. Biological Aspects
of the Link Between Smoking and Depression. Harvard Rev. Psychiatry
2000, 8, 99−110.
(23) Philip, N. S.; Carpenter, L. L.; Tyrka, A. R.; Price, L. H.
Nicotinic Acetylcholine Receptors and Depression: A Review of the
Preclinical and Clinical Literature. Psychopharmacology (Berlin, Ger.)
2010, 212, 1−12.
(24) Andreasen, J. T.; Nielsen, E. O.; Christensen, J. K.; Olsen, G.
M.; Peters, D.; Mirza, N. R.; Redrobe, J. P. Subtype-selective Nicotinic
Acetylcholine Receptor Agonists Enhance the Responsiveness to
Citalopram and Reboxetine in the Mouse Forced Swim Test. J.
Psychopharmacology 2011, 25, 1347−1356.
(25) Andreasen, J. T.; Redrobe, J. P. Nicotine, but not Mecamyl-
amine, Enhances Antidepressant-like Effects of Citalopram and
Reboxetine in the Mouse Forced Swim and Tail Suspension Tests.
Behav. Brain Res. 2009, 197, 150−156.
(7) In comparison DHβE has a Ki of 98 nM at the α4β2 nAChR and
a Ki of 10.5 μm at the α7 subtype.
(8) Decker, M. W.; Anderson, D. J.; Brioni, J. D.; Donnellyroberts, D.
L.; Kang, C. H.; Oneill, A. B.; Piattonikaplan, M.; Swanson, S.;
Sullivan, J. P. Erysodine, a Competitive Antagonist at Neuronal
Nicotinic Acetylcholine Receptors. Eur. J. Pharmacol. 1995, 280, 79−
89.
(9) (a) Bergmeier, S. C.; Ismail, K. A.; Arason, K. M.; McKay, S.;
Bryant, D. L.; McKay, D. B. Structure−Activity Studies of Ring E
Analogues of Methyllycaconitine. Part 2: Synthesis of Antagonists to
α3β4* Nicotinic Acetylcholine Receptors Through Modifications to
the Ester. Bioorg. Med. Chem. Lett. 2004, 14, 3739−3742. and
references cited therein. (B) For a recent example at the α4β2 nAChR
subtype, see: German, N.; Kim, J.-S.; Jain, A.; Dukat, M.; Pandya, A.;
Ma, Y.; Weltzin, M.; Schulte, M. K.; Glennon, R. A. Deconstruction of
the α4β2 Nicotinic Acetylcholine Receptor Positive Allosteric
Modulator Desformylflustrabromine. J. Med. Chem. 2011, 54, 7259−
7267.
(10) (a) Hider, R. C.; Walkinshaw, M. D.; Saenger, W. Erythrina
Alkaloid Nicotinic Antagonists: Structure−Activity Relationships. Eur.
J. Med. Chem. 1986, 21, 231−234. (b) Megirian, D.; Leary, D. E.;
Slater, I. H. The Action of some Derivatives of beta-Erythroidine on
Peripheral Neuro-effector Systems. J. Pharmacol. Exp. Ther. 1955, 113,
212−227.
9681
dx.doi.org/10.1021/jm4013592 | J. Med. Chem. 2013, 56, 9673−9682

Journal of Medicinal Chemistry
Article
(26) Popik, P.; Kozela, E.; Krawczyk, M. Nicotine and Nicotinic
Receptor Antagonists Potentiate the Antidepressant-like Effects of
Imipramine and Citalopram. Br. J. Pharmacol. 2003, 139, 1196−1202.
(27) Caldarone, B. J.; Harrist, A.; Cleary, M. A.; Beech, R. D.; King, S.
L.; Picciotto, M. R. High-Affinity Nicotinic Acetylcholine Receptors
are Required for Antidepressant Effects of Amitriptyline on Behavior
and Hippocampal Cell Proliferation. Biol. Psychiatry 2004, 56, 657−
664.
(28) Andreasen, J. T.; Olsen, G. M.; Wiborg, O.; Redrobe, J. P.
Antidepressant-like Effects of Nicotinic Acetylcholine Receptor
Antagonists, but not Agonists, in the Mouse Forced Swim and
Mouse Tail Suspension Tests. J. Psychopharmacol. 2009, 23, 797−804.
(29) Kristensen, A. S.; Andersen, J.; Jørgensen, T. N.; Sørensen, L.;
Eriksen, J.; Loland, C. J.; Strømgaard, K.; Gether, U. SLC6
Neurotransmitter Transporters: Structure, Function, and Regulation.
Pharmacol. Rev. 2011, 63, 585−640.
(30) Andersen, J.; Kristensen, A. S.; Bang-Andersen, B.; Strømgaard,
K. Recent advances in the understanding of the interaction of
antidepressant drugs with serotonin and norepinephrine transporters.
Chem. Commun. 2009, 3677−3692.
(31) Menachery, M. D.; Lavanier, G. L.; Wetherly, M. L.;
Guinaudeau, H.; Shamma, M. Simple Isoquinoline Alkaloids. J. Nat.
Prod. 1986, 49, 745−778.
(32) Bembenek, M. E.; Abell, C. W.; Chrisey, L. A.; Rozwadowska,
M. D.; Gessner, W.; Brossi, A. Inhibition of Monoamine Oxidases A
and B by Simple Isoquinoline Alkaloids: Racemic and Optically Active
1,2,3,4-Tetrahydro-, 3,4-Dihydro-, and Fully Aromatic Isoquinolines. J.
Med. Chem. 1990, 33, 147−152.
(33) Shulman, K. I.; Herrmann, N.; Walker, S. E. Current Place of
Monoamine Oxidase Inhibitors in the Treatment of Depression. CNS
Drugs 2013, 27, 789−797.
(34) (a) Slattery, D. A.; Cryan, J. F. Using the Rat Forced Swim Test
to Assess Antidepressant-like Activity in Rodents. Nature Protoc. 2012,
7 (6), 1009−1014. (b) Can, A.; Dao, D. T.; Arad, M.; Terrillion, C. E.;
Piantadosi, S. C.; Gould, T. D. The Mouse Forced Swim Test. J. Vis.
Exp. 2012, 29, e3638.
(35) Schneider, W.; Muller, B. Beitrage zur Chemie der Carbinol-
̈
̈
amine. I Untersuchungen an Py-Tetrahydroisochinolinen. Justus
Liebigs Ann. Chem. 1958, 615, 34−42.
(36) Ahmed-Schofield, R.; Mariano, P. S. A Photochemical Route for
Erythrane Ring Construction. J. Org. Chem. 1987, 52, 1478−1482.
(37) (a) Armishaw, C.; Jensen, A. A.; Balle, T.; Clark, R. J.; Harpsøe,
K.; Skonberg, C.; Liljefors, T.; Strømgaard, K. Rational Design of α-
Conotoxin Analogues Targeting α7 Nicotinic Acetylcholine Receptors:
Improved Antagonistic Activity by Incoporation of Proline Derivatives.
J. Biol. Chem. 2009, 284, 9498−9512. (b) Hansen, C. P.; Jensen, A. A.;
Christensen, J. K.; Balle, T.; Liljefors, T.; Frølund, B. Novel
acetylcholine and carbamoylcholine analogues: development of a
functionally selective alpha4beta2 nicotinic acetylcholine receptor
agonist. J. Med. Chem. 2008, 51, 7380−7395.
(38) Sørensen, L.; Andersen, J.; Thomsen, M.; Hansen, S. M.; Zhao,
X.; Sandelin, A.; Strømgaard, K.; Kristensen, A. S. Interaction of
Antidepressants with the Serotonin and Norepinephrine Transporters:
Mutational Studies of the S1 Substrate Binding Pocket. J. Biol. Chem.
2012, 287, 43694−43707.
9682
dx.doi.org/10.1021/jm4013592 | J. Med. Chem. 2013, 56, 9673−9682