Exploration of the molecular architecture of the orthosteric binding site in the α4β2 nicotinic acetylcholine receptor with analogs of 3-(dimethylamino)butyl dimethylcarbamate (DMABC) and 1-(pyridin-3-yl)-1,4-diazepane

Article abstract of DOI:10.1016/j.ejmech.2015.07.024

X-ray crystal structures of acetylcholine binding proteins (AChBPs) have revealed two different possible extensions to the classical ligand binding pocket known to accommodate various nicotinic agonists. One of the pockets is limited in size while the oth

Full text of DOI:10.1016/j.ejmech.2015.07.024

Accepted Manuscript
Exploration of the Molecular Architecture of the Orthosteric Binding Site in the
α4β2 Nicotinic Acetylcholine Receptor with Analogs of 3-(Dimethylamino)butyl
Dimethylcarbamate (DMABC) and 1-(Pyridin-3-yl)-1,4-diazepane
Tinna B. Bach, Anders A. Jensen, Jette G. Petersen, Troels E. Sørensen, Serena
Della Volpe, Jun Liu, Antoni R. Blaazer, Jacqueline E. van Muijlwijk-Koezen, Thomas
Balle, Bente Frølund
PII:
S0223-5234(15)30156-2
10.1016/j.ejmech.2015.07.024
EJMECH 8008
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 24 February 2015
Revised Date: 10 July 2015
Accepted Date: 13 July 2015
Please cite this article as: T.B. Bach, A.A. Jensen, J.G. Petersen, T.E. Sørensen, S. Della Volpe,
J. Liu, A.R. Blaazer, J.E. van Muijlwijk-Koezen, T. Balle, B. Frølund, Exploration of the Molecular
Architecture of the Orthosteric Binding Site in the α4β2 Nicotinic Acetylcholine Receptor with Analogs
of 3-(Dimethylamino)butyl Dimethylcarbamate (DMABC) and 1-(Pyridin-3-yl)-1,4-diazepane, European
Journal of Medicinal Chemistry (2015), doi: 10.1016/j.ejmech.2015.07.024.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Graphical abstract
1

ACCEPTED MANUSCRIPT
Exploration of the Molecular Architecture of the
Orthosteric Binding Site in the α4β2 Nicotinic
Acetylcholine Receptor with Analogs of 3-
(Dimethylamino)butyl Dimethylcarbamate (DMABC) and
1
-(Pyridin-3-yl)-1,4-diazepane
#
#
#
#,†
Tinna B. Bach, Anders A. Jensen, Jette G. Petersen, Troels E. Sørensen, Serena
#
†
Della Volpe, Jun Liu, Antoni R. Blaazer,º Jacqueline E. van Muijlwijk-Koezen,º
†
,
#,
Thomas Balle, * and Bente Frølund *
#
Department of Drug Design and Pharmacology, Faculty of Health and Medical
Sciences, University of Copenhagen, Universitetsparken 2, DK-2100 Copenhagen,
Denmark
†
Faculty of Pharmacy, The University of Sydney, Sydney, NSW 2006, Australia
ºDivision of Medicinal Chemistry, Amsterdam Institute for Molecules, Medicines and
Systems, Faculty of Sciences, VU University Amsterdam, De Boelelaan 1105, 1081 HV
Amsterdam, The Netherlands
*
Corresponding Authors:
Bente Frølund, e-mail: bfr@sund.ku.dk, phone: (+45)35336495
and Thomas Balle, e-mail: thomas.balle@sydney.edu.au, phone (+61)290367035
1

ACCEPTED MANUSCRIPT
ABSTRACT
X-ray crystal structures of Acetylcholine binding proteins (AChBPs) have revealed two
different possible extensions to the classical ligand binding pocket known to
accommodate various nicotinic agonists. One of the pockets is limited in size while the
other is of considerable dimensions and protrudes along the interfacial cleft between
subunits. To probe these putative extensions in functional nicotinic acetylcholine
receptors (nAChRs), elongated analogs of 3-(dimethylamino)butyl dimethylcarbamate
(DMABC) and 1-(pyridine-3-yl)-1,4-diazepane were prepared and characterized
pharmacologically at neuronal heteromeric nAChRs. Although the new analogs, relative
to parent compounds, displayed lower binding affinities, functional characterization of
selected compounds revealed that they had retained partial α4β2 nAChR agonist
activity. The structure-activity relationship data did not indicate an upper limit to the
size of substituents as would have been expected if the bound in the smaller pocket. The
data were better in agreement with a binding mode in which substituents protrude along
the interfacial cleft of the receptor. This was further supported by docking into a
homology model of the α4-β2 nAChR interface and by surface plasmon resonance
biosensor analysis of binding of the compounds to acetylcholine-binding proteins,
where they exhibit preference for Lymnaea stagnalis ACh binding protein (Ls-AChBP)
over the Aplysia california ACh binding protein (Ac-AChBP). These results suggest
new opportunities for expanding chemical space in the development of partial agonist
and may be of interest in relation to development of novel smoking cessation aids.
Keywords: nACh receptor agonists, structure-activity studies, Ls-AChBP, Ac-AChBP,
Lobeline, DMABC analogs
1
. Introduction
2

ACCEPTED MANUSCRIPT
The neurotransmitter acetylcholine (ACh, 1, Figure 1) mediates its physiological effects
in the central and peripheral nervous systems through two classes of receptors: the
muscarinic and nicotinic ACh receptors (mAChRs and nAChRs, respectively) [1-4].
Central nAChRs are involved in regulation of release of other neurotransmitters,
including dopamine, norepinephrine, serotonin, glutamate, and γ-aminobutyric acid [5-
8
]. The fact that imbalances in the cholinergic system have been proposed to be
involved in devastating diseases such as schizophrenia, depression, epilepsy, and
Alzheimer’s and Parkinson’s diseases has spawned the interest in nAChRs as putative
drug targets in these disorders [1, 3].
Neuronal nAChRs consist of five subunits that can be either identical (homomeric, α
only) or different (heteromeric, combination of α and β subunits) [1][12]. The α4β2 and
α7 receptors are the most abundant nAChRs in the brain while α3b4 nAChRs dominate
in the autonomic ganglia. [1, 13]. The orthosteric binding sites in the heteromeric
(+) (-)
receptors are located at the extracellular α /β interface (α and β contributing with the
principal and complementary binding component, respectively).
ACh binding proteins (AChBPs) have proven valuable as surrogate proteins for
structural studies of orthosteric ligand binding to nAChRs [16-20]. The AChBP is a
soluble homopentameric protein complex with a structure and fold similar to that of the
extracellular domain of the nAChR. Although the overall sequence identities between
AChBPs and nAChRs are low (20-24%), the interfacial ACh binding site in the AChBP
is highly conserved including the “aromatic box” which consists of five aromatic
residues. These residues, TyrA, TrpB, TyrC1, TyrC2 and TrpD, named according to the
loop in the orthosteric nAChR site on which they reside [21] (Figure 2A), are strictly
conserved among all nAChRs as well as Ls-AChBP. In Ct- and Ac-AChBP TrpD in the
nAChR is replaced by an isoleucine and a tyrosine, respectively. From more than 50
3

ACCEPTED MANUSCRIPT
AChBP co-crystal structures, it is clear that the binding site is flexible, and that the
aromatic box residues play an important role in defining its shape and size. When an
agonist binds, as illustrated by (S)-nicotine bound to Ls-AChBP [24] (Figure 2A), the
aromatic residues pack around the ligand to form a closed binding site. In response to
more bulky ligands, including antagonists methyllycaconitine [17], α-cobratoxin [25],
and α-conotoxin ImI [26], TyrC1 and TyrC2 on the flexible loop C which forms a lid on
the binding site can easily be displaced outwards. Two other aromatic box residues
observed to adopt alternative conformation in AChBP structures are TyrA and TrpD
(TyrD in Ac-AChBP). Alternative conformations of these residues have revealed two
different extensions to the binding pocket that can accommodate non-classical ligands,
including lobeline [17, 20] (2, Figure 1 and Figure 2B) and the synthetic compound, 3α-
(benzoyloxy)-8β-((R)-2-hydroxy-2-phenylethyl)-8α-methyl-8
azoniabicyclo[3.2.1]octane methiodide (3, Figure 1 and Figure 2C) [27]. From
crystallographic studies of Ac-AChBP it is known that a sub-pocket is made accessible
by rotameric shift of TyrA upon binding of 2 (Figure 2B) [17]. However, mutational
studies indicate that the flip is dependent on specific residues on Loop F on the
complementary subunit interface to stabilize the tyrosine in its alternative conformation
[
27], and it remains unknown if the pocket exists in nAChRs. The second sub-pocket is
an extension of the binding site along the dimer interface towards the ion channel and
has been observed by co-crystallization of 3 with Ac-AChBP [27]. Simultaneous
repositioning of TyrA and TyrD in Ac-AChBP exposes a narrow cleft where the ligand
is wedged between the two aromatic residues (Figure 2C). The existence of this
interfacial pocket is also evident from co-crystal structures with MLA and from
covalent trapping experiments with a maleimide analog of MLA in both α7 and α4β2
receptors [28, 29], it is evident that the pocket is not an AChBP artifact. The alternative
4

ACCEPTED MANUSCRIPT
sub-pockets as extensions to the classical agonist binding site have so far received little
attention.
In previous studies, we have explored the medicinal chemistry of 3-
(dimethylamino)butyl dimethylcarbamate (4, DMABC) and its desmethyl analog 5 [31-
3
3], and 1-(pyridin-3-yl)-1,4-diazepane (6) [34] as potent nAChR agonists (Figure 1).
Co-crystal structures with Ls-AChBP have revealed that both 4 and 6 form direct
contacts to TyrA, the residue observed to adopt a flipped state in the Ac-AChBP co-
crystal structure with 2 [35, 36] (SI Figure 1A,B). Inspired by these co-crystal
structures, we have in the present study elaborated on the scaffolds of 4 and 6 to explore
possible extensions of the agonist binding pocket and to investigate the structure-
activity relationships (SARs) of elongated analogs of these structures at the α4β2
nAChR.
2
. Results and discussion
2
.1 Ligand design
Based on overlays of AChBP co-crystal structures with 4 and 6 on co-crystal structures
with 2 and 3 (Figures 3A-D), we designed a series of ligands with substituents that
could protrude towards the two cavities (illustrated in Figures 2B and 2C). In the first
series, one of the methyl groups on the basic amine of 4 was replaced with 2-hydroxy-
phenethyl and phenylpropyl substituents to give compounds 7 and 8 (Table 1). These
compounds were designed to challenge the position of TyrA potentially allowing access
to the sub-pocket associated with the binding of 2 (Figure 2B). Along these lines, a
series of compounds based on bioisosteric replacements of the dimethylcarbamate
moiety with methylcarbamate to give compounds 9–12 (Table 1) was also planned. In a
second series, the 3-methyl group of 4 was replaced with longer arylalkyl substituents
from phenethyl to phenylpentyl to give compounds 13–16 and aryl ethers 17–21 (Table
5

ACCEPTED MANUSCRIPT
1
). These compounds were inspired from the overlay of 4 and 3 (Figure 3B) indicating
that large substituents in this position could protrude along the interfacial cleft (Figure
C) without compromising the binding mode of the core agonist fragment. Again,
2
bioisosteric replacements of the dimethylcarbamate with methylcarbamate and an
imidazolyloxy moiety were explored resulting in C3 substituted analogs 22–27 (Table
1
). In a third series, we explored the 1,4-diazepane scaffold of 6. As N-4 forms a
hydrogen bond to TyrA in the X-ray structure, substitution in this position could lead to
either tyrosine flip exposing the subpocket associated with binding of 2 in Ac-AChBP
or alternatively a minor change in binding mode allowing the substituent to exit the
binding site via the cavity along the interfacial cleft (Figure 3C and 3D). Thus benzyl,
phenethyl, phenylpropyl, and phenpropenyl substituents were introduced on the N-4 of
the 1,4-diazepane to challenge these areas in extension to the orthosteric binding site
and resulted in compounds 28–33 (Table 2). In the analog series of 6, we also explored
a series of piperazine analogs, 34–36, with benzyl, phenethyl and phenylpropyl
substituents on N-4 of the piperazine ring (Table 2).
2
.2 Chemistry
The syntheses of N-arylalkyl analogs of 4 (compounds 7–12) are illustrated in Scheme
. The synthesis of 9 and 10 was achieved by a three-step procedure starting from
methyl acetoacetate and either methyl benzylamine or methyl phenethylamine to yield
8 and 39. Alcohols 40 and 41 were obtained by reduction with sodium which after
1
3
treatment with 1,1’-carbonyldiimidazole (CDI) and either methyl- or dimethylamine (in
either H O or ethanol) gave 9, 10 and 37. Products 9 and 37 were exposed to
2
hydrogenolysis that afforded 42 and 43 in good yields which subsequently were reacted
with either the appropriate epoxide to give 7 and 11 or subjected to reductive amination
to afford 8 and 12.
6

ACCEPTED MANUSCRIPT
The synthesis approaches to C-3-arylalkyl analogs of 4 are presented in Schemes 2 and
3
. Target compounds 13–15 and 22–25 were synthesized starting from 44–47 (Scheme
2
), which have been previously described [37, 38]. Subsequent Michael addition
afforded compounds 48–51 that were treated with LiAlH to give the alcohols 52–55.
4
Carbamate formations were performed as described above to yield the final compounds.
Compound 27 was obtained by reacting the alkoxide of 54 with 2,4,5-tribromo-1-
methyl-1H-imidazole and subsequent treatment of the product with n-BuLi (Scheme 2).
Target compounds 17 and 18 were synthesized from 56[31] by an Ullmann like
reaction, whereas 19-21 were obtained by nucleophilic aromatic substitution with 56 as
the nucleophile (Scheme 3). A different approach was applied for the synthesis of the C-
3
-arylalkyl analogs 16 and 26 to introduce the hydroxyl group in the α position to the
phenyl ring (Scheme 3). Alcohol 57[39] was protected as the tetrahydropyranyl (THP)
ether using our previously published procedure [31]. The procedures for the following
steps to afford 16 and 26 are described above, and the final target compounds 16 and 26
were obtained after THP-deprotection with p-toluenesulfonic acid monohydrate.[31]
The syntheses of the N-arylalkyl 3-pyridyl diazacycle analogs are presented in Scheme
4
. Analog 28 was synthesized in two steps from 58[40]. First, intermediate 59 was
obtained by a selective bromination in the 6-position of 58 using NBS. Subsequent Boc-
deprotection followed by reductive amination afforded the analog. Compounds 29–36
were synthesized as described for 28. All target compounds were characterized
pharmacologically as the oxalate salts.
7

ACCEPTED MANUSCRIPT
2
.3 Pharmacology and SAR discussion
The binding affinities displayed by the target compounds and reference compounds 1, 2,
3
and 4-6 in a [ H]-epibatidine competition binding assays to membranes from HEK293
cells stably expressing recombinant rat α4β2, α4β4, and α3β4 nAChRs are presented in
Tables 1 and 2. Functional data for selected compounds and reference compounds
determined in a fluorescence-based functional assay the FLIPR Membrane Potential
Blue assay and in two-electrode-voltage-clamp (TEVC) electrophysiology recordings at
nAChR expressed in Xenopus oocytes are given in Tables 3 and 4. Dissociation
constants determined for 30 and 32 at Ac- and Ls-AChBP by surface plasmon resonance
(
SPR) biosensor analyses, respectively, are shown in Table 5.
In the following, SARs have been classified into groups: N-arylalkyl analogs of 4
compounds 7–12); C-3-arylalkyl analogs of 4 (compounds 13–26); and N-arylalkyl 3-
pyridyl diazacycle analogs (compounds 28–36).
(
2
.3.1 N-arylalkyl analogs of DMABC (4)
Earlier reported SAR studies for analogs of 4 showed that incorporating the amino
group into a piperidine or pyrrolidine ring decreased binding affinities significantly
relative to 4 [33]. Based on overlay of the X-ray structure of AChBP co-crystallised
with structures 2 and 4 (Figure 3A) this can be rationalized by either a loss of affinity
due to the energetic penalty associated with moving TyrA away to allow access to the
sub-pocket associated with binding of 2 or, alternatively, by an altered binding mode of
the core agonist fragment required to position the increased steric bulk in the sub-pocket
associated with binding of 3. We hypothesized that if the first scenario was in play we
should be able to alter the size of the substituent on the basic nitrogen to optimally fit
the sub-pocket illustrated in Figure 2B. This should then be reflected in improved
8

ACCEPTED MANUSCRIPT
binding affinities up until a certain point after which it should drop drastically. For the
second scenario a more flat SAR is to be expected. The resulting N-arylalkyl analogs 7–
1
2 exhibited binding affinities in the mid-to-high micromolar range corresponding to a
reduction in affinity at α4β2 nAChRs of more than 900-fold compared to 4. No
significant recovery of affinity was observed with increasing size of substituents, nor
upon incorporation of a hydroxyl group (7 and 11) to mimic the hydroxyl group in 2.
Together, this SAR does not suggest that substituents access the subpocket analogous to
the one observed upon binding of 2 in Ac-AChBP.
2
.3.2 C3-arylalkyl analogs of DMABC (4)
We have previously reported that presence of a small hydrophobic group, preferably a
methyl group, at the C-3 carbon of 4 was optimal for nAChR receptor binding and have
found that the (R)-enantiomer of 4 is significantly more potent than the (S)-enantiomer
[
33]. Introduction of larger hydrophobic substituents, including isopropyl or cyclohexyl,
in this position induced a shift from nanomolar to micromolar affinity at the heteromeric
nAChRs [33]. Introduction of the more planar cyclopropyl, 3-thiophenyl, and phenyl
substituents also led to significant decreases in binding affinity compared to the parent
compound, 4 [31]. However, overlay of the X-ray structures of 3 and 4 (Figure 3B)
suggested that even larger substituents in the C-3 position of 4 could protrude along the
interfacial cleft basically without limitation to length of the substituents. Therefore,
arylalkyl chains with varying length were introduced in the C-3 position to probe the
effect of larger substituents. The resulting compounds (13–15) exhibited affinities in the
low micromolar range corresponding to 45–95 fold loss in affinity relative to 4. For the
monomethylcarbamates (22–25) a decrease in affinities of more than 800-fold relative
to 5 was observed. Interestingly, elongation of the alkyl chain from two to five carbons
did not affect binding affinity significantly as evident comparing the phenethyl
9

ACCEPTED MANUSCRIPT
substituted analog 22 (7.1 µM, α4β2 nAChR) to the phenylpentyl substituent of 25 (2.7
µM, α4β2 nAChR). A similar trend was observed for the dimethylcarbamate analogs
1
3–15. This indicates that the binding pocket can accommodate substituents with a
significant difference in length and is compatible with the flat SAR that was expected
for compounds protruding into an interfacial cleft (Figure 2C).
In a previous study, imidazolyloxy moities were shown to be carbamate bioisosteres at
the α4β2 nAChRs [35]. Exchange of the carbamate in compound 24 for a 1-methyl-1H-
imidazoloxy moiety (27) yielded the most potent compound of the series with an
affinity at the α4β2 nAChR of 0.76 µM and a selectivity for the α4β2 nAChRs over the
α4β4 and α3β4 nAChRs, in concordance with data for the parent compounds [35].
To illustrate compatibility of the elongated ligands with a structural model of the
receptor, compounds 13–15, 22–25, and 27 were docked into a homology model of the
α4β2 nAChR interface constructed using the co-crystal structure of AChBP with 3 to
ensure a model with an accessible interfacial cleft. The suggested binding modes of the
R-enantiomers of these compounds only differed with respect to how far along the
interfacial cleft the substituent protruded (Figure 4A). In addition, the binding mode
was found not to compromise the binding mode of the parent agonist fragment, 4, which
in a co-crystal structure was identified as the (R)-enantiomer [35]. Analogs of 4 with
aryl ethers in the C3-position (17–21) were found to have decreased affinities relative to
4
and corresponding alkyl phenyl analogs with substituents of comparable size as
evident by the 14- and 3-fold lower affinity (α4β2 nAChR) of 17 and 20 relative to 22
and 23. These data are in concordance with observations made from previous C-3
substituted analogs of 4 in which introduction of a hydroxymethyl group in the C-3
position led to a significantly larger decrease in nAChR binding affinity compared to
4
,[31] whereas the binding affinity of the corresponding ethyl analog was much less
affected.[41] These data are consistent with a high internal energy of 17 (~10 kJ/mol)
1
0

ACCEPTED MANUSCRIPT
for the conformation resembling the bioactive conformation of 4 [35]. Internal hydrogen
bonding from the protonated amine to the ether oxygen is preferred as opposed to the
internal hydrogen bond to the carbamate oxygen observed in the bound conformation of
4
(SI Figure 1). As the compounds originally were inspired by 2, we investigated the
effect of a hydroxyl group in the α-position relative to the phenyl ring. In 2, this is
highly favorable and leads to a 25-fold increase in affinity [42]. In the present series the
effect was opposite and led to ca. 10-fold decrease in affinity at α4β2 nAChRs
(compare 13 and 23 to 16 and 26).
2
.3.3 N-arylalkyl 3-pyridyl diazacyle analogs
3
-Pyridyl substituted diazacycles and diazabicycles are privileged structures and have
been the prime scaffolds for synthesis of high affinity α4β2 nAChR agonists. Among
these, the scaffold of 6 has been extensively applied in the study of nAChRs in which
introduction of various substituents in the 5- and 6-positions on the pyridine ring
afforded high-affinity compounds. Especially, rather large substituents including
methoxyalkyl groups were allowed in the 5-position but not in the 6-position.
Furthermore, analogs of 6 have been reported in which the 1,4-diazepane ring was
exchanged for the piperazine ring which generally led to a reduction in α4β2 affinities
by up to two orders of magnitude [34]. To our knowledge, the nAChRs have not
systematically been challenged by large substituents in the N-4 position of the scaffold
of 6 and the corresponding piperazine scaffold.
The overlay of X-ray structures of 2 and 6 (Figure 3C) suggested an arylalkyl chain
length of approximately 1-2 carbons in the N-4 position of 6 to be optimal for reaching
into the subpocket associated to the binding of 2. On the other hand, a chain length of
three should be too long. Alternatively, if accessing the interfacial cleft pocket, arylalkyl
1
1

ACCEPTED MANUSCRIPT
substituents could in principle be of any length but would require a slight change in
binding mode of the core agonist fragment to allow the substituent to escape the cavity.
Introduction of phenylalkyl substituents (benzyl, phenethyl, and phenylpropyl) led to
compounds 30–32 which displayed 100–150 fold reduced binding affinities at α4β2
nAChRs compared to the parent compound, 6. Since 6 is a highly potent α4β2 nAChR
agonist (Table 2), the N-4 substituted analogs still displayed binding affinities in the low
to medium nanomolar range. The equivalent series of 1-(pyridin-3-yl)piperazine
analogs (34–36) displayed 5- to 10-fold lower affinity at the three heteromeric nAChRs
compared to the 1,4-diazepane containing analogs 30–32. However, a similar SAR
pattern was observed for analogs of 6 and the corresponding piperazine analogs
indicating a similar binding mode of the analogs in which the benzyl, phenethyl, and
phenylpropyl substituents in the N-4 positions are tolerated with a flat SAR in
agreement with binding along the interfacial cleft as illustrated by docking in Figure 4B.
Based on docking experiments, the analogs 30–32 are positioned less optimal for
interaction of the core agonist fragment and in particular the protonated N-4 with the
receptor. This may explain the relatively larger decrease in affinities (>100-fold)
compared to the C-3 substituted analogs of 4 which only suffered a ca. 35-fold
reduction in affinity and which upon docking had its core agonist part optimally
positioned as in the X-ray structure. To probe the importance of reduction of entropic
penalty of binding, an alkenyl analog of 32, 29, was evaluated. However, 32 and 29
were found equipotent at the receptors indicating that this specific reduction in
conformational flexibility is not important for receptor binding. In an attempt to
increase binding affinity, a bromine atom was inserted in the 6-position of the pyridine
ring of 6, which previously was found to improve affinity and potency [34, 43].
However, the analog (28) exhibited a 10-fold decrease in affinity at the α4β2 nAChRs
1
2

ACCEPTED MANUSCRIPT
compared to 32 which again may be indicative of a slightly shifted binding mode of the
core agonist fragment. Introduction of bromine in the meta-position of the phenyl ring
of 31 gave compound 33 which was equipotent to 31 at the α4β2 receptors but more
than 4-fold more potent at the α3β4 nAChRs compared to 31 indicating room for
improvement of affinity in this specific position.
2
.3.4 Functional characterization
Selected compounds of the C-3-arylalkyl analogs of 4 and the N-arylalkyl 3-pyridyl
diazacycle analogs of 6 (Table 3) were characterized functionally at recombinant
heteromeric mouse α4β2 and rat α3β4 nAChRs in the FLIPR Membrane Potential Blue
assay. Compounds 8, 12, 17, and 20 were devoid of activity at concentrations up to 300
µM (not shown). The results revealed all remaining compounds to be weak agonists
except for 29, which exhibited weak antagonistic properties (Table 3). The
pharmacological activity of compounds 30-33 could not be determined in the FLIPR
Membrane Potential Blue assay. One of these, 30, was characterized functionally by
two-electrode voltage clamp electrophysiological recordings on α4β2 receptors
expressed in Xenopus laevis oocytes. On oocytes injected with α4 and β2 mRNA in a
1
:4 ratio to favor assembly of (α4) (β2) receptors[15] a partial agonist response
2 3
corresponding to ca 16% of the maximal response of ACh and an EC value of 1.8 µM
5
0
were observed. On oocytes injected with α4 and β2 mRNA in a 10:1 ratio favoring
expression of (α4) (β2) receptors the response was 6% of the maximal ACh elicited
3
2
current and a slightly higher EC of 7.9 µM. (Table 4 and Figure 5).
5
0
The agonistic properties of 13–15, 23, 24, 27, 30, 35, and 36 are highly interesting and
support the docking studies suggesting that the substituted analogs have a binding mode
1
3

ACCEPTED MANUSCRIPT
in which the core agonist fragment of 4 and 6 at least to some degree is conserved
because the compounds are still able to act as agonists.
2
.3.5 Characterization of compounds 30 and 32 at Ac- and Ls-AChBP
Since the target compounds were hypothesized to address cavities based on studies of
the Ac-AChBP and Ls-AChBP, we included the two AChBPs in the profiling of benzyl
and phenylpropyl N-4 substituted analogs of 6 (30 and 32) using surface plasmon
resonance (SPR) biosensor analyses (Table 5). The reference compounds (S)-nicotine
and 2 were included in the study, and the data for these compounds were in agreement
with data previously reported [27]. The results revealed 30 and 32 to have ~6- and ~50-
fold higher affinity for Ls- over Ac-AChBP, respectively. Being equipotent at the α4β2
nAChRs, 32 (K = 0.020 µM) showed an 8-fold higher affinity at Ls-AChBP compared
d
to 30 (K = 0.16 µM) but despite this deviation the binding affinities obtained for
d
compounds 30 and 32 to Ls-AChBP reflected the binding data at α4β2 nAChRs (Table
5
). This correlation corresponds well to previous observations made from analogs of
4
[35] and 6[36] in which high-affinity compounds at the α4β2 receptors were found to
hold good affinity at Ls-AChBP. Further, as the hydrophobic cavity associated with
binding of 2 in Ac-AChBP has been suggested not to be present in Ls-AChBP it
supports our conclusion of a binding mode of elongated nAChR agonists along the
interfacial cleft of α4β2 nAChRs.
2
.4 Conclusion
N- and C-3-arylalkyl analogs of 4 and N-arylalkyl analogs of 6 were designed and
synthesized to explore possible extensions to the nAChR agonist binding pocket. Based
on X-ray structures of AChBP with 2 and 3, two such pockets were potentially
accessible but the flat SAR for elongated analogs of the compounds investigated here is
1
4

ACCEPTED MANUSCRIPT
best in agreement with binding in one of these pockets that protrudes along the
interfacial cleft as observed in the co-crystal structure of 6 with Ac-AChBP. To be
compatible with binding in the hydrophobic lobeline pocket, a clear optimum in terms
of binding affinities for compounds with intermediate size substituents was expected
but not observed. The suggested binding mode is further supported by docking and SPR
analysis on AChBPs which show preference for Ls-AChBP over Ac-AChBP. These
findings propose a new opportunity for design of partial agonists with properties like 2,
and the fact that agonistic properties were obtained with using both 4 and 6 as starting
point suggests that this type of ligands could be generally accessible from agonist
scaffolds.
3
. Experimental section
3
3
.1 Computational chemistry
.1.1 Homology modeling
A homology model of the α4β2 nAChR interface was constructed as described by
Harpsøe et al.[15] with the following exceptions: The Ac-AChBP crystal structure used
as template was 2Y58[27] instead of 2BYQ[17]. This change necessitated changes in
the alignment to accommodate minor differences in sequence between the two
templates. These changes included the removal of H1 and S2 in the beginning of the
sequence from the alignment (said residues are not present in 2Y58), and changing A43
and A138 to valine (these residues are mutated to valine in 2Y58).
3
.1.2 Ligand docking
Initally, the N-4 benzyl substituted 6 was docked to the model using the induced fit
docking protocol implemented in Schrodinger’s Maestro v9.3. Default settings were
1
5

ACCEPTED MANUSCRIPT
used except ligand scaling was set to 0.8 and amino acid residues within 8Å of the
docked ligand were sampled. The highest ranking pose with an agonist-like binding
mode, i.e. hydrogen bond from pyridine nitrogen to water molecule and hydrogen atom
of the protonated N-4 of 1,4-diazepane pointing towards the backbone carbonyl oxygen
of Trp143, was selected. In subsequent docking runs the Glide v5.8[44] program was
used together with the Glide extra precission (XP) scoring function. The binding modes
and docking scores were selected based on ranking according to the Glide E-model
score.
3
3
.2. Chemistry
.2.1. General procedures
All reactions involving air sensitive reagents were performed under a nitrogen
atmosphere using syringe-septum cap techniques. When using dry solvents, diethyl
ether (Et O) was stored over sodium wires, and THF was distilled or obtained from a
2
solvent purification system. Furthermore, dry DCM and DMF were either obtained by
drying over molecular sieves (4Å) or from a solvent purification system. All other
solvents, stated dry, were purchased from a commercial source and stored over
1
13
molecular sieves (4Å). H NMR and C NMR spectra were obtained on a Varian
Gemini 2000 (300 MHz), Varian Mercury Plus (300 MHz), or Bruker Avance (400
1
MHz). Data are tabulated in the following order for H NMR: chemical shift (δ)
(multiplicity (bs, broad singlet; s, singlet; d, doublet; dd, doublet of doublets; dq, dublet
of quartets; ddd, doublet of doublet of doublets; dt, doublet of triplets; ddt, doublet of
doublet of triplets; dtd, doublet of triplet of doublets; p, pentet; q, quartet; qd, quartet of
doublets; t, triplet; td, triplet of doublets; tt, triplet of triplets; m, multiplet; sxt, sextet;
sep, septet), coupling constant(s) J (Hz), number of protons). The solvent residual peak
1
3
was used as internal reference. Carbon-fluorine coupling constants observed in
C
1
6

ACCEPTED MANUSCRIPT
NMR of fluorine containing compounds are stated. Analytical thin-layer
chromatography (TLC) was carried out using Merck silica gel 60 F254 plates, and the
compounds were visualized with UV light (254 nm) and/or KMnO spraying reagent.
4
Flash column chromatography (FCC) was performed using Merck silica gel 60 (0.040–
0
.063 mm). Dry column vacuum chromatography (DCVC) was performed using Merck
silica gel 60 (0.015–0.040 mm). Reverse phase column chromatography was performed
using Merck LiChroprep RP-18 (0.040–0.063 mm). Liquid chromatography-mass
spectrometry (LC-MS) was performed using an Agilent 1200 series system connected to
a Bruker Esquire 3000Plus Iontrap MS system with a gradient of water + 0.2% formic
acid (Buffer A) and MeCN + 0.2% formic acid (Buffer B). All microwave reactions
were carried out in a glass vial using a Biotage Initiator instrument. Elemental analyses
were performed by Johannes Theiner, Department of Physical Chemistry, University of
Vienna, Austria, and are within ± 0.4% of the theoretical values, unless otherwise
stated. Melting points (mp) were determined by OptiMelt from Stanford Research
Systems in open capillary tubes and are uncorrected.
3
.2.2. General procedure for oxalate salt formation
The free amine (1.0 eq) was dissolved in a small amount of warm acetone to which a
solution of oxalic acid (1.1 eq) in warm acetone was added. The mixture was stored at 5
°
C until crystallization. If crystallization did not occur a few drops of Et O was added.
2
3
.2.3. 13-((2-Hydroxy-3-phenylpropyl)(methyl)amino)butyl dimethylcarbamate (7)
To a solution of 43 (200 mg, 1.15 mmol) in MeCN (5 mL), LiClO (24 mg, 0.23 mmol)
4
and 2-benzyloxirane (185 mg, 1.38 mmol) were added, and the reaction mixture was
stirred at reflux overnight. The reaction mixture was diluted with EtOAc (40 mL) and
washed with water (30 mL). The water phase was extracted with EtOAc (2 × 40 mL).
1
7

ACCEPTED MANUSCRIPT
The combined organic phases were dried (MgSO ), filtered, and evaporated in vacuo.
4
Purification of the crude product was obtained by DCVC (Heptane:EtOAc:NH /100:0:0
3
1
to 0:99:1) yielding the product as a clear oil (218 mg, 61%). H NMR (CDCl , 300
3
MHz) δ 7.39-7.14 (m, 5H), 4.23-4.07 (m, 2H), 3.98-3.82 (m, 1H), 3.05-2.76 (m, 8H),
2
.68 (dd, 1H, J = 13.6, 5.6 Hz), 2.56-2.12 (m, 5H), 1.73-1.54 (m, 1H), 1.36-1.24 (m,
H), 1.03 (s, 3H). 3-((2-Hydroxy-3-phenylpropyl)-(methyl)amino)butyl
1
dimethylcarbamate oxalate. The free amine of 7 (218 mg, 0.71 mmol) was converted
into the oxalate salt as described under the general procedure affording white crystals
1
(
39 mg, 14%). Mp: n.d. H NMR (D O, 300 MHz) δ 7.32-7.09 (m, 5H), 4.20-3.87 (m,
2
3
H), 3.63-3.38 (m, 1H), 3.10-2.91 (m, 2H), 2.82-2.66 (m, 12H), 2.21-1.57 (m, 2H), 1.25
13
(
dd, 3H, J = 11.1, 6.7 Hz). C NMR (D O, 75 MHz) δ 165.64 (2C); 157.93, 157.76
2
(diastereomers); 136.76, 136.69 (diastereomers); 129.52 (2C), 129.39, 128.82 (2C,
diastereomers); 127.05; 67.27, 67.09, 66.35, 66.25 (diastereomers); 62.32, 61.88, 61.79,
1.02, 60.52 (diastereomers); 58.57, 57.79, 56.69 (diastereomers); 56.47, 56.09, 54.94
diastereomers); 40.90, 40.74 (2C, diastereomers); 38.01, 37.28, 36.87 (diastereomers);
6
(
3
6.70, 36.13, 35.74 (diastereomers); 31.17, 31.06, 29.23, 27.95 (diastereomers); 14.27,
3.99, 12.18, 11.65 (diastereomers).
1
3
.2.4. 3-[Phenylpropyl(methyl)amino]butyl dimethylcarbamate (8)
To a solution of 43 (200 mg, 1.15 mmol) and 3-phenylpropionaldehyde (0.15 mL, 1.15
mmol) dissolved in 1,2-dichloroethane (10 mL) and MeOH (10 mL) was added AcOH
(
0.07 mL, 1.15 mmol) and NaBH CN (109 mg, 1.75 mmol). The reaction mixture was
3
stirred at room temperature for 72 hours. The reaction mixture was quenched by adding
saturated aqueous NaHCO solution (20 mL) and was then extracted with EtOAc (3 ×
3
3
0 mL). The organic phases were combined, dried (MgSO ), filtered, and evaporated in
4
vacuo. The crude product was purified by DCVC (Toluene:DCM/100:0 to 95:5)
1
8

ACCEPTED MANUSCRIPT
yielding a pale yellow oil. The product was purified further by extraction: The product
was poured into water (30 mL) and then acidified with aqueous HCl solution (30 mL, 2
M), followed by extraction with EtOAc (30 mL). The water phase was then basified
with aqueous NaOH solution (30 mL, 2M) and extracted with EtOAc (2 × 60 mL). The
combined organic phases were dried (MgSO ), filtered, and evaporated in vacuo
4
1
yielding the product as a pale yellow oil (127 mg, 38%). H NMR (CDCl3, 300 MHz) δ
7
2
3
8
.30-7.08 (m, 5H), 4.12-4.00 (m, 2H), 2.85 (s, 6H), 2.70-2.53 (m, 2H), 2.45-2.15 (m,
H), 2.18 (s, 3H), 1.98-1.61 (m, 4H), 1.60-1.41 (m, 1H), 0.99, 0.97 (2 s, 3H, rotamers).
-[Phenylpropyl(methyl)amino]butyl dimethylcarbamate oxalate. The free amine of
(104 mg, 0.36 mmol) was converted into the oxalate salt as described under the
1
general procedure affording white crystals (28 mg, 27%). Mp: 107.5-108.3 °C. H NMR
(
(
(
D O, 400 MHz) δ 7.44-7.34 (m, 2H), 7.33-7.26 (m, 3H), 4.24-4.13 (m, 1H), 4.13-4.00
2
m, 1H), 3.62-3.50 (m, 1H), 3.23-2.99 (m, 2H), 2.86, 2.84 (2 s, 6H, rotamers), 2.79-2.67
1
3
m, 5H), 2.15-1.71 (m, 4H), 1.34, 1.28 (2 d, 3H, J = 6.8 Hz). C NMR (D O, 101 MHz)
2
δ 165.54 (2C); 157.78, 157.74 (rotamers); 140.49, 140.42 (rotamers); 128.84 (2C);
1
3
28.47; 128.46; 126.58; 61.83 (2C); 58.14, 57.94 (rotamers); 53.29, 52.36 (rotamers);
5.70, 35.14 (rotamers); 31.74 (2C); 30.72; 28.56, 25.74 (rotamers); 13.79, 11.83
(rotamers).
3
.2.5. 3-[Benzyl(methyl)amino]butyl methylcarbamate (9)
A solution of 40 (626 mg, 3.24 mmol) in dry toluene (9 mL) was added CDI (632 mg,
3
.90 mmol), stirred for 3 h at rt, and added dry THF (3.3 mL). After further stirring for
2
.5 h, methylamine (0.37 mL, 7.0 mmol, 40% in H O) was added. The reaction mixture
2
was stirred for 17 h, diluted with EtOAc (9 mL), and washed with brine (8 mL). The
organic phase was dried (MgSO ), filtered, and evaporated in vacuo. Purification by
4
DCVC (DCM:MeOH:NH /100:0:0 to 300:4.5:0.5) afforded the product as a yellow oil
3
1
9

ACCEPTED MANUSCRIPT
1
(
572 mg, 71%). H NMR (CDCl , 300 MHz) δ 7.34-7.18 (m, 5H), 4.48 (bs, 1H), 4.18 (t,
3
2
2
3
1
H, J = 6 Hz), 3.59 (d, 1H, J = 12 Hz), 3.46 (d, 1H, J = 12 Hz), 2.89-2.82 (m, 1H), 2.80,
.78 (2 s, 3H, rotamers), 2.14 (s, 3H), 1.91-1.84 (m, 1H), 1.65-1.56 (m, 1H), 1.02 (d,
1
3
H, J = 9 Hz). C NMR (CDCl , 75 MHz) δ 157.8, 140.6, 129.1 (2C), 128.5 (2C),
3
27.1, 63.4, 58.2, 54.5, 36.5, 33.6, 27.9, 13.4. 3-[Benzyl(methyl)amino]butyl
methylcarbamate oxalate. The free amine of 9 (470 mg, 1.89 mmol) was converted
into the oxalate salt as described under the general procedure affording white crystals
1
(
383 mg, 49%) after recrystallization from acetone. Mp: 118.0-119.8 °C. H NMR
(
D O, 300 MHz) δ 7.43-7.40 (m, 5H), 4.32 (dd, 1H, J = 12, 3 Hz), 4.21-4.13 (m, 1H),
2
4
.09-3.98 (m, 2H), 3.52-3.38 (m, 1H), 2.68 (s, 3H), 2.61, 2.57 (2 s, 3H, rotamers), 2.45-
1
3
2
.38, 2.28-2.18, 2.09-1.99, 1.90-1.79 (4 m, 2H), 1.34, 1.32 (2 d, 3H, J = 6 Hz). C
NMR (D O, 75 MHz) δ 165.8 (2C); 159.0; 130.9 (2C); 130.4; 129.6 (2C); 61.4; 57.7,
2
5
2
7.5 (rotamers); 57.0, 56.5 (rotamers), 35.8, 35.1 (rotamers); 31.0, 29.1 (rotamers);
6.9; 13.9, 12.0 (rotamers).
3
.2.6. 3-[Phenethyl(methyl)amino]butyl methylcarbamate (10)
Compound 10 was prepared as described for 9 from a mixture of 41 (2.96 g, 14.3
mmol), CDI (2.76 g, 17.0 mmol), and methylamine (1.65 mL, 29 mmol, 40% in H O).
2
Purification by DCVC (DCM:MeOH:NH /100:0:0 to 100:4.5:0.5) afforded the product
3
1
as a yellow oil (2.95 g, 80%). H NMR (CDCl , 300 MHz) δ 7.22-7.07 (m, 5H), 4.48
3
(bs, 1H), 3.99 (t, 2H, J = 6 Hz), 2.78-2.43 (m, 5H), 2.71, 2.70 (2 s, 3H, rotamers), 2.18
(s, 3H), 1.72 (dq, 1H, J = 12, 6 Hz), 1.45 (dq, 1H, J = 12, 6 Hz), 0.88 (d, 3H, J = 6 Hz).
1
3
C NMR (CDCl , 75 MHz) δ 157.7, 141.1, 129.1 (2C), 128.7 (2C), 126.3, 63.4, 55.8,
3
5
5.4, 37.0, 35.5, 33.4, 31.4, 13.82. 3-[Phenethyl(methyl)amino]butyl
methylcarbamate oxalate. The free amine of 10 (100 mg, 0.38 mmol) was converted
into the oxalate salt as described under the general procedure affording white crystals
2
0

ACCEPTED MANUSCRIPT
1
(
103 mg, 77%). Mp: 141.1-142.3 °C. H NMR (D O, 300 MHz) δ 7.36-7.25 (m, 5H),
2
4
2
3
.13-3.95 (m, 2H), 3.62-3.22 (m, 3H), 3.08-2.89 (m, 2H), 2.79, 2.75 (2 s, 3H, rotamers),
.62, 2.60 (2 s, 3H, rotamers), 2.10-1.95 (m, 1H), 1.89-1.74 (m, 1H), 1.29, 1.26 (2 d,
1
3
H, J = 6 Hz). C NMR (D O, 75 MHz) δ 165.9 (2C); 159.0; 136.2; 129.4 (2C); 129.0
2
(2C); 127.8; 61.8, 61.5 (rotamers); 59.1, 58.5 (rotamers); 54.8, 53.4 (rotamers); 36.6,
3
5.5 (rotamers); 30.7, 30.5 (rotamers); 29.4; 26.9; 13.6, 12.0 (rotamers).
3
.2.7. 3-((2-Hydroxy-2-phenethyl)(methyl)amino)butyl methylcarbamate (11)
Compound 11 was prepared as described for 7 from a mixture of 42 (944 mg, 5.93
mmol), LiClO (125 mg, 1.17 mmol), and styrene oxide (850 mg, 7.07 mmol).
4
Purification by HPLC (MeOH:H O (pH 10)/40:60) yielded the product as a clear oil (84
2
1
mg, 58%). H NMR (CDCl , 300 MHz) δ 7.40-7.22 (m, 5H), 4.79 (bs, 1H), 4.70-4.64
3
(
m, 1H), 4.20-4.11 (m, 2H), 2.96-2.74 (m, 1H), 2.81, 2.79 (2 d, 3H, J = 6 Hz), 2.65, 2.54
2 dd, 1H, J = 12, 3 Hz), 2.43, 2.39 (2 dd, 1H, J = 12, 3 Hz), 2.34, 2.27 (2 s, 3H,
(
1
3
rotamers), 1.93-1.75 (m, 1H), 1.68-1.58 (m, 1H), 1.00 (d, 3H, J = 6 Hz). C NMR
(
CDCl , 75 MHz) δ 157.6; 142.8; 128.7 (2C); 127.8; 126.2 (2C); 69.8, 69.4
3
(diastereomers); 64.7 (diastereomers); 63.2, 63.0 (diastereomers); 60.9 (diastereomers);
5
7.4, 55.6 (diastereomers); 37.9 (diastereomers); 34.11; 33.7, 33.5 (diastereomers);
2
7.8; 14.5, 13.7 (diastereomers). 3-((2-Hydroxy-2-phenethyl)(methyl)amino)butyl
methylcarbamate oxalate. The free amine of 11 (65 mg, 0.23 mmol) was converted
into the oxalate salt as described under the general procedure affording white crystals
1
(
27 mg, 22%). Mp: 80.5-82.9 °C. H NMR (D O, 300 MHz) δ 4.98-4.94 (m, 1H), 4.06-
2
3
.93 (m, 2H), 3.61-3.41 (m, 1H), 3.35-3.11 (m, 2H), 2.84, 2.79, 2.68 (3 s, 3H,
diastereomers), 2.52, 2.50 (2 s, 3H, diastereomers), 2.13-1.95 (m, 1H), 1.84-1.61 (m,
1
3
1
H), 1.26-1.16 (m, 3H). C NMR (D O, 75 MHz) δ 166.0 (2C); 159.0; 139.5; 129.1
2
(2C); 126.3 (2C); 126.2; 68.4, 67.4 (diastereomers); 62.0, 61.8 (diastereomers); 61.3;
2
1

ACCEPTED MANUSCRIPT
5
1
8.1; 38.1, 36.4 (diastereomers); 31.0, 29.5 (diastereomers); 28.9, 26.9 (diastereomers);
3.7, 12.4 (diastereomers).
3
.2.8. 3-[Phenylpropyl(methyl)amino]butyl methylcarbamate (12)
Compound 12 was prepared as described for 8 from a mixture of 42 (200 mg, 1.25
mmol), 3-phenylpropionaldehyde (0.16 mL, 1.25 mmol), and NaBH CN (109 mg, 1.75
3
1
mmol) affording the product as a yellow oil (104 mg, 30%). H NMR (CD OD, 300
3
MHz) δ 7.40-7.10 (m, 5H), 4.53 (s, 1H), 4.14 (m, 2H), 2.90-2.70 (m, 4H), 2.69-2.51
(m, 2H), 2.49-2.28 (m, 2H), 2.18 (s, 3H), 1.92-1.66 (m, 3H) 1.64-1.43 (m, 1H), 0.90,
0
.93 (2 s, 3H, rotamers). 3-[Phenylpropyl-(methyl)amino]butyl methylcarbamate
oxalate. The free amine of 12 (104 mg, 0.37 mmol) was converted into the oxalate salt
as described under the general procedure affording white crystals (52 mg, 22%). Mp:
1
1
4
21.5-123.3 °C. H NMR (D O, 400 MHz) δ 7.43-7.34 (m, 2H), 7.33-7.24 (m, 3H),
2
.25-4.14 (m, 1H), 4.12- 3.99 (m, 1H), 3.61-3.48 (m, 1H), 3.22-2.97 (m, 2H), 2.80-2.62
13
(
m, 8H), 2.15-1.95 (m, 3H), 1.93-1.70 (m, 1H), 1.33, 1.27 (2 d, 3H, J = 6.5 Hz). C
NMR (D O, 101 MHz) δ 165.53 (2C); 140.50, 140.45 (rotamers); 128.84 (2C); 128.49;
2
1
28.47; 126.58; 118.92; 61.31; 58.12; 53.31; 35.11; 31.74; 30.61; 28.81, 26.64
(rotamers); 25.74, 25.72 (rotamers); 13.60, 11.81 (rotamers).
3
.2.9. 3-(Dimethylamino)-6-phenylhexyl dimethylcarbamate (13)
A solution of 53 (1.52 g, 6.87 mmol) in dry toluene (25 mL) was added CDI (1.34 g,
.24 mmol). The reaction mixture was stirred for 2 h at rt. The reaction mixture was
8
added THF (20 mL) and stirred for additional one hour. Dimethylamine (8.70 mL, 68.7
mmol, 40% in water) was added the reaction mixture, and stirring continued overnight
at rt. The reaction mixture was diluted with EtOAc (25 mL). The water phase was
washed with brine (25 mL) and aqueous NaOH solution (1M, 2 × 25 mL). The
2
2

ACCEPTED MANUSCRIPT
combined organic phases were dried (MgSO ), filtered, and evaporated yielding the
4
1
product as a yellow oil (1.54 g, 77%). H NMR (CDCl , 400 MHz) δ 7.29-7.27 (m, 1H),
3
7
.25 (d, 1H, J = 1.8 Hz), 7.18-7.16 (m, 3H), 4.10 (d, 2H, J = 13.7 Hz), 2.89 (s, 6H), 2.61
(t, 2H, J = 7.6 Hz), 2.52 (t, 1H, J = 5.8 Hz), 2.23 (s, 6H), 1.82 (dq, 1H, J = 13.6, 6.8
1
3
Hz), 1.70-1.64 (m, 2H), 1.57 (t, 2H, J = 6.9 Hz), 1.35-1.26 (m, 1H). C NMR (CDCl ,
3
1
01 MHz) δ 156.86, 142.54, 128.52 (2C), 128.41 (2C), 125.84, 63.91, 61.00, 40.41
(4C), 36.20, 29.48, 29.10, 28.95. 3-(Dimethylamino)-6-phenylhexyl
dimethylcarbamate oxalate. The free amine of 13 (380 mg, 1.30 mmol) was converted
into the oxalate salt as described under the general procedure affording white crystals
1
(
291 mg, 59%). Mp: 97.9 °C. H NMR (D O, 400 MHz) δ 7.34-7.29 (m, 2H), 7.24-7.20
2
(m, 3H), 4.11 (ddd, 1H, J = 11.5, 6.6, 4.9 Hz), 4.02 (ddd, 1H, J = 11.7, 7.4, 4.5 Hz),
3
.28-3.22 (m, 1H), 2.84 (s, 6H), 2.79, 2.77 (2 s, 6H, rotamers), 2.69-2.61 (m, 2H), 2.06
13
(
ddt, 1H, J = 15.1, 7.6, 4.9 Hz), 1.93-1.85 (m, 1H), 1.73-1.61 (m, 4H). C NMR (D O,
2
1
01 MHz) δ 165.63 (2C), 157.67, 141.74, 128.69 (2C), 128.55 (2C), 126.23, 63.55
(2C), 62.28, 39.54, 38.77, 34.36 (2C), 27.72, 27.66, 26.70.
3
.2.10. 3-(Dimethylamino)-7-phenylheptyl dimethylcarbamate (14)
Compound 14 was prepared as described for 13 from a mixture of 54 (800 g, 3.40
mmol), CDI (660 g, 4.08 mmol), and dimethylamine (4.3 mL, 34 mmol, 33% in EtOH)
1
yielding the product as a yellow oil (860 mg, 64%). H NMR (CDCl , 400 MHz) δ
3
7
.29-7.27 (m, 2H), 7.17 (dt, 3H, J = 1.4, 0.7 Hz), 4.12 (t, 2H, J = 6.9 Hz), 2.90 (s, 6H),
.61 (t, 2H, J = 7.8 Hz), 2.49-2.47 (m, 1H), 2.24 (s, 6H), 1.64-1.61 (m, 2H), 1.59-1.52
2
(m, 2H), 1.42-1.26 (m, 4H). 3-(Dimethylamino)-7-phenylheptyl dimethylcarbamate
oxalate. The free amine of 14 (367 mg, 1.20 mmol) was converted into the oxalate salt
as described under the general procedure affording white crystals (346 mg, 94%). Mp:
1
9
0.4 °C. H NMR (D O, 400 MHz) δ 7.42-7.37 (m, 2H), 7.33-7.27 (m, 3H), 4.22 (ddd,
2
2
3

ACCEPTED MANUSCRIPT
1
5
1
7
H, J = 11.5, 6.5, 5.1 Hz), 4.13 (ddd, 1H, J = 11.7, 7.4, 4.5 Hz), 3.33 (tt, 1H, J = 7.6,
.1 Hz), 2.90 (s, 6H), 2.80, 2.79 (2 s, 6H, rotamers), 2.70 (t, 2H, J = 7.4 Hz), 2.15 (ddt,
H, J = 15.0, 7.6, 5.0 Hz), 2.02-1.93 (m, 1H), 1.82-1.68 (m, 4H), 1.43 (dt, 2H, J = 15.3,
1
3
.6 Hz). C NMR (D O, 101 MHz) δ 165.65 (2C), 157.68, 142.65, 128.60 (2C), 128.54
2
(2C), 125.98, 63.57, 62.20, 39.42 (2C), 38.94 (2C), 34.51, 30.05, 28.18, 27.77, 24.58.
3
.2.11. 3-(Dimethylamino)-8-phenyloctyl dimethylcarbamate (15)
Compound 15 was prepared as described for 13 from a mixture of 55 (488 mg, 1.96
mmol), CDI (381 mg, 2.35 mmol), and dimethylamine (1.75 mL, 9.78 mmol, 33% in
1
EtOH) giving the product as a yellow oil (332 mg, 53%). H NMR (CD OD, 400 MHz)
3
δ 7.27-7.19 (m, 2H), 7.18-7.10 (m, 3H), 4.10 (td, 2H, J = 6.7, 1.9 Hz), 2.89 (s, 6H), 2.61
(t, 2H, J = 7.8 Hz), 2.48 (p, 1H, J = 6.3 Hz), 2.22 (s, 6H), 1.84 (dq, 1H, J = 14.1, 6.3
1
3
Hz), 1.67-1.51 (m, 4H), 1.42-1.22 (m, 6H). C NMR (CD OD, 101 MHz) δ 144.04,
3
1
3
30.08, 129.56 (2C), 129.41 (2C), 126.80, 65.23, 62.32, 40.85 (2C), 37.00, 36.32 (2C)
2.78, 30.74, 30.53, 30.45, 28.13. 3-(Dimethylamino)-8-phenyl-octyl
dimethylcarbamate oxalate. The free amine of 15 (128 mg, 0.40 mmol) was converted
into the oxalate salt according to the general procedure affording a white solid (64 mg,
1
3
9%). Mp: 84.7-86.0 °C. H NMR (D O, 400 MHz) δ 7.38-7.32 (m, 2H), 7.30-7.21 (m,
2
3
H), 4.23-4.07 (m, 2H), 3.27 (d, 1H, J = 6.8 Hz), 2.87, 2.87 (2 s, 6H, rotamers), 2.80
(dd, 6H, J = 6.3, 1.5 Hz), 2.66-2.59 (m, 2H), 2.17-2.06 (m, 1H), 2.01-1.88 (m, 1H),
1
3
1
1
3
.76-1.57 (m, 4H), 1.44-1.28 (m, 4H). C NMR (D O, 101 MHz) δ 165.64 (2C),
2
52.37, 143.16, 128.59 (2C), 128.58 (2C), 125.88. 63.53, 62.31, 39.49, 38.87, 35.94,
5.56, 34.84, 34.82, 30.20, 28.28, 27.74, 24.87.
3
.2.12. 3-(Dimethylamino)-6-hydroxy-6-phenylhexyl dimethylcarbamate (16)
2
4

ACCEPTED MANUSCRIPT
To a solution of 61 (0.92 g, 2.34 mmol) in MeOH (50 mL) was added p-toluenesulfonic
acid monohydrate (0.89 g, 4.68 mmol). The reaction mixture was stirred for 2 h at rt.
The reaction mixture was added saturated aqueous NaHCO solution (50 mL) and then
3
reduced in vacuo. The remaining mixture was extracted with DCM (3 × 50 mL) and the
combined organic phases were dried, filtered, and evaporated in vacuo. The crude
product was purified by DCVC (DCM:MeOH:NH /90:9:1) yielding the product as a
3
1
yellow oil (90 mg, 13%). H NMR (CDCl , 400 MHz) δ 7.38-7.19 (m, 5H), 4.81 (dd,
3
1
H, J = 6.0, 3.6 Hz), 4.16-4.10 (m, 1H), 4.05-3.99 (m, 1H), 2.89 (s, 6H), 2.56 (tt, 1H, J
9.7, 3.3 Hz), 2.33, 2.25 (2 s, 6H, rotamers), 2.17-2.09 (m, 1H), 1.97-1.87 (m, 2H),
=
1
3
1
1
3
.60-1.52 (m, 1H), 1.45-1.30 (m, 2H). C NMR (CDCl , 101 MHz) δ 156.66, 145.80,
3
28.21 (2C), 126.59, 125.74 (2C), 72.17, 63.56, 61.51, 40.08 (4C), 37.21, 27.29, 26.73.
-(Dimethylamino)-6-hydroxy-6-phenylhexyl dimethylcarbamate oxalate. The free
amine of 16 (90 mg, 0.29 mmol) was converted into the oxalate salt as described under
1
the general procedure affording white crystals (67 mg, 58%). Mp: 119.3 °C. H NMR
(
D O, 400 MHz) δ 7.51-7.40 (m, 5H), 4.81 (d, 1H, J = 6.3 Hz), 4.20 (ddd, 1H, J = 11.5,
2
6
.8, 4.9 Hz), 4.12 (ddd, 1H, J = 11.7, 7.3, 4.6 Hz), 3.35 (td, 1H, J = 6.2, 2.1 Hz), 2.89 (s,
1
3
6
H), 2.85, 2.80 (2 s, 6H, rotamers), 2.21-2.13 (m, 1H), 2.04-1.68 (m, 5H). C NMR
(
D O, 101 MHz) δ 165.69 (2C), 157.66, 142.95, 128.82 (2C), 128.08, 126.02 (2C),
2
7
3.01, 63.41, 62.23, 39.47 (2C), 38.87 (2C), 33.63, 27.71, 24.73.
3
.2.13. 3-(Dimethylamino)-4-phenoxybutyl dimethylcarbamate (17)
5
6[31] (200 mg, 0.98 mmol), CuI (6 mg, 0.03 mmol), 3,4,7,8-tetramethyl-1,10-
phenanthroline (15 mg, 0.07 mmol), iodobenzene (73 µL, 0.65 mmol), and Cs CO (319
2
3
mg, 0.98 mmol) were mixed and added dry toluene (0.50 mL) under nitrogen. The
reaction mixture was heated to 110 °C overnight, diluted with EtOAc (15 mL), and
filtered through a plug of silica washing additional times with EtOAc. The filtrate was
2
5

ACCEPTED MANUSCRIPT
1
concentrated in vacuo yielding the product as a brown oil (136 mg, 60%). H NMR
(
CDCl , 300 MHz) δ 7.38-7.28 (m, 2H), 7.05-6.88 (m, 3H), 4.22 (td, 2H, J = 6.6, 1.9
3
Hz), 4.17-4.05 (m, 1H), 4.02-3.96 (m, 1H), 3.05-3.01 (m, 1H), 2.92, 2.89 (2 s, 6H,
rotamers), 2.43 (s, 6H), 1.99-1.87 (m, 3H), 1.73-1.56 (m, 3H). 3-(Dimethylamino)-4-
phenoxybutyl dimethylcarbamate oxalate. The free amine of 17 (185 mg, 0.66 mmol)
was converted into the oxalate salt as described under the general procedure affording
1
white crystals (66 mg, 27%). Mp: 91.3-92.5 °C. H NMR (D O, 300 MHz) δ 7.40-7.35
2
(m, 2H), 7.09-7.00 (m, 3H), 4.46 (dd, 1H, J = 11.9, 3.1 Hz), 4.31 (dd, 1H, J = 11.9, 6.6
Hz), 4.23 (dd, 2H, J = 9.8, 5.7 Hz), 3.84-3.81 (m, 1H), 2.93, 2.88 (2 s, 6H, rotamers),
1
3
2
1
3
.77 (s, 6H), 2.27 (m, 2H). C NMR (D O, 75 MHz) δ 241.85 (2C), 215.94, 186.69,
2
44.11 (2C), 136.32, 128.79 (2C), 78.13, 77.16, 76.48, 55.46, 53.41, 50.18, 49.74,
8.60.
3
.2.14. 3-(Dimethylamino)-4-(4-fluorophenoxy)butyl dimethylcarbamate (18)
Compound 18 was prepared as described for 17 from a mixture of 56 (200 mg, 0.98
mmol), CuI (6 mg, 0.03 mmol), 3,4,7,8-tetramethyl-1,10-phenanthroline (16 mg, 0.07
mmol), 1-fluoro-4-iodobenzene (76 µL, 0.66 mmol), and Cs CO (320 mg, 0.98 mmol).
2
3
Purification
by reverse phase
column chromatography (C18 silica)
(
H O:AcOH:MeCN/99.9:0.1:0 to 94.9:0.1:5) followed by basic extraction of the
2
1
product into EtOAc afforded the product as a yellow oil (46 mg, 23%). H NMR
(
CD OD, 300 MHz) δ 7.04-6.90 (m, 3H), 4.19 (t, 2H, J = 6.5 Hz), 4.07 (qd, 2H, J =
3
1
3
1
0.3, 5.0 Hz), 3.01-2.97 (m, 1H), 2.88 (s, 5H), 2.37 (s, 6H), 2.01-1.91 (m, 2H). C
1
NMR (CD OD, 75 MHz) δ 160.06, 158.11 (C-F: J = 147 Hz); 156.92; 156.06, 156.03
3
4
2
3
(
C-F: J = 2.1 Hz); 116.72, 116.59 (C-F: J = 8.5 Hz, 2C); 116.49, 116.41 (C-F: J = 5.8
Hz, 2C); 68.59; 64.57; 61.29; 41.61 (2C); 36.57; 36.13; 28.22. 3-(Dimethylamino)-4-
4-fluorophenoxy)butyl dimethylcarbamate oxalate. The free amine of 18 (46 mg,
(
2
6

ACCEPTED MANUSCRIPT
0
.15 mmol) was converted into the oxalate salt as described under the general procedure
1
affording white crystals (43 mg, 72%). Mp: 97.6-98.3 °C. H NMR (D O, 300 MHz) δ
2
7
.07-7.00 (m, 2H), 6.97-6.90 (m, 2H), 4.39 (dd, 1H, J = 11.8, 3.1 Hz), 4.27-4.12 (m,
H), 3.78 (qd, 1H, J = 6.7, 3.0 Hz), 2.87, 2.82 (2 s, 6H, rotamers), 2.68, 2.66 (2 s, 5H,
4
1
3
rotamers), 2.18-2.26 (m, 2H). C NMR (D O, 75 MHz) δ 165.59; 159.18, 156.04 (C-F:
2
1
3
2
J = 245 Hz); 157.52; 153.29, 153.26 (C-F: J = 2.3 Hz, 2C); 116.19, 115.80 (C-F: J =
2
2
1 Hz); 115.88, 115.69 (C-F: J = 21 Hz); 64.48; 62.97; 62.24; 41.42; 39.16; 35.96;
3
5.52; 24.36.
3
.2.15. 3-(Dimethylamino)-4-(pyridin-2-yloxy)butyl dimethylcarbamate (19)
To a suspension of NaH (65 mg, 1.62 mmol, 60% in mineral oil) in dry DMF (6 mL)
was dropwisely added 56 (300 mg, 1.47 mmol) at 0 °C. After 1 h of stirring at 0 °C, 2-
fluoropyridine (0.25 mL, 2.94 mmol) was added. The reaction mixture was stirred
overnight at rt, added water (60 mL), and extracted with Et O (2 × 60 mL). The
2
combined organic phases were dried (MgSO ), filtered, and evaporated in vacuo.
4
Purification
by reverse phase
column chromatography (C18 silica)
(
H O:TFA:MeCN/99.9:0.1:0 to 89.9:0.1:10) followed by basic extraction of the product
2
1
into EtOAc afforded the product as a yellow oil (224 mg, 54%). H NMR (CD OD, 300
3
MHz) δ 8.15 (dd, 1H, J = 5.1, 1.3 Hz), 7.75 (ddd, 1H, J = 8.4, 7.1, 1.9 Hz), 7.04 (ddd,
1
H, J = 7.1, 5.2, 1.0 Hz), 6.93 (d, 1H, J = 8.4 Hz), 4.80 (dd, 1H, J = 13.1, 3.0 Hz), 4.63
(dd, 1H, J = 13.1, 6.6 Hz), 4.54-4.50 (m, 1H), 4.31-4.24 (m, 2H), 3.86 (qd, 1H, J = 6.5,
1
3
3
1
3
.3 Hz), 3.02 (s, 6H), 2.86 (s, 6H), 2.34-2.16 (m, 2H). C NMR (CD OD, 75 MHz) δ
3
63.25, 157.57, 147.31, 140.95, 119.18, 112.10, 63.82, 63.12, 62.83, 41.19 (2C), 36.68,
6.15, 26.30. 3-(Dimethylamino)-4-(pyridin-2-yloxy)butyl dimethylcarbamate
oxalate. The free amine of 19 (224 mg, 0.80 mmol) was converted into the oxalate salt
as described under the general procedure affording white crystals (141 mg, 47%). Mp:
2
7

ACCEPTED MANUSCRIPT
1
7
8
1
2
7.1 °C. H NMR (D O, 300 MHz) δ 8.08 (dd, 1H, J = 5.2, 1.7 Hz), 7.77 (ddd, 1H, J =
2
.6, 7.1, 1.9 Hz), 7.11-7.07 (m, 1H), 6.94 (dd, 1H, J = 8.4, 0.6 Hz), 4.66 (dd, 1H, J =
2.6, 2.7 Hz), 4.50 (dd, 1H, J = 12.6, 6.1 Hz), 4.26-4.15 (m, 2H), 3.87-3.81 (m, 1H),
1
3
.90 (s, 6H), 2.68, 2.65 (2 s, 5H, rotamers), 2.29-2.17 (m, 2H). C NMR (D O, 75
2
MHz) δ 166.09, 161.69 (2C), 157.48, 145.72, 141.37, 118.72, 111.00, 62.85, 62.71,
6
2.33, 36.02 (2C), 35.51 (2C), 24.62.
3
.2.16. 4-(Benzyloxy)-3-(dimethylamino)butyl dimethylcarbamate (20)
Compound 20 was prepared as described for 19 from a mixture of NaH (44 mg, 1.08
mmol, 60% in mineral oil), 56 (200 mg, 0.98 mmol), and benzyl bromide (0.23 mL,
1
1
.96 mmol) yielding the product as a yellow oil (94 mg, 33%). H NMR (CDCl , 300
3
MHz) δ 7.38-7.27 (m, 5H), 4.53 (s, 2H), 4.14 (t, 2H, J = 6.7 Hz), 3.62-3.47 (m, 2H),
.92, 2.89 (2 s, 7H, rotamers), 2.37 (s, 6H), 1.92-1.70 (m, 2H). 4-(Benzyloxy)-3-
dimethylamino)butyl dimethylcarbamate oxalate. The free amine of 20 (94 mg,
2
(
0
.32 mmol) was converted into the oxalate salt as described under the general procedure
1
affording white crystals (114 mg, 93%). Mp: 79.0-79.7 °C. H NMR (D O, 300 MHz) δ
2
7
3
2
1
3
.40-7.34 (m, 5H), 4.58 (d, 2H, J = 0.7 Hz), 4.11-3.95 (m, 2H), 3.77 (dd, 1H, J = 12.0,
.7 Hz), 3.65 (dd, 1H, J = 12.0, 8.0 Hz), 3.53-3.41 (m, 1H), 2.83-2.73 (m, 12H), 2.18-
13
.06 (m, 1H), 1.92-1.85 (m, 1H). C NMR (D O, 75 MHz) δ 165.64, 157.52 (2C),
2
36.70, 128.85 (2C), 128.69 (2C), 128.62, 73.14, 65.32, 63.14, 62.20, 41.22, 38.02,
6.08, 35.65, 24.02.
3
.2.17. 3-(Dimethylamino)-4-(thiazol-2-yloxy)butyl dimethylcarbamate (21)
Compound 21 was prepared as described for 19 from a mixture of NaH (65 mg, 1.62
mmol, 60% in mineral oil), 56 (300 mg, 1.47 mmol), and 2-bromothiazole (0.26 mL,
2
.94 mmol). Purification by reverse phase column chromatography (C18 silica)
2
8