Conformationally restrained carbamoylcholine homologues. Synthesis, pharmacology at neuronal nicotinic acetylcholine receptors and biostructural considerations

Article abstract of DOI:10.1016/j.ejmech.2015.07.029

Exploration of small selective ligands for the nicotinic acetylcholine receptors (nAChRs) based on acetylcholine (ACh) has led to the development of potent agonists with clear preference for the α₄β₂ nAChR, the most prevalent nAChR subtype in the central nervous system. In this work we present the continuation of these efforts aimed at increasing this subtype selectivity by introduction of conformational restriction in the carbamoylcholine homologue, 3-(dimethylaminobutyl) dimethylcarbamate (DMABC). Our results highlight the importance of the N-carbamoyl substitution in α₄β₂-subtype selectivity. Moreover, we have confirmed the non-linear conformation of DMABC bound to nAChRs suggested by recent crystal structures of the compound in complex with the Lymnaea stagnalis ACh binding protein.

Full text of DOI:10.1016/j.ejmech.2015.07.029

Accepted Manuscript
Conformationally restrained carbamoylcholine homologues. Synthesis, pharmacology
at neuronal nicotinic acetylcholine receptors and biostructural considerations
Mario de la Fuente Revenga, Thomas Balle, Anders A. Jensen, Bente Frølund
PII:
S0223-5234(15)30161-6
10.1016/j.ejmech.2015.07.029
EJMECH 8013
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 10 March 2015
Revised Date: 2 July 2015
Accepted Date: 16 July 2015
Please cite this article as: M.d.l. Fuente Revenga, T. Balle, A.A. Jensen, B. Frølund, Conformationally
restrained carbamoylcholine homologues. Synthesis, pharmacology at neuronal nicotinic acetylcholine
receptors and biostructural considerations, European Journal of Medicinal Chemistry (2015), doi:
10.1016/j.ejmech.2015.07.029.
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

ACCEPTED MANUSCRIPT
Title: Conformationally restrained carbamoylcholine
homologues. Synthesis, pharmacology at neuronal
nicotinic acetylcholine receptors and biostructural
considerations.
1
2
3
Authors: Mario de la Fuente Revenga , Thomas Balle , Anders A. Jensen ,
3
,*
Bente Frølund .
Authors filiation:
1
Instituto de Química Médica, Consejo Superior de Investigaciones Científicas
(
IQM-CSIC), C/Juan de la Cierva 3, 28006-Madrid, Spain.
2
Faculty of Pharmacy. The University of Sydney, NSW, Australia.
University of Copenhagen, Universitetsparken 2, DK-2100 Copenhagen,
3
Denmark.
*
Corresponding author: Bente Frølund, Phone: +45 35336495. E-mail:
bfr@sund.ku.dk
1

ACCEPTED MANUSCRIPT
ABSTRACT.
Exploitation of small selective ligands for the nicotinic acetylcholine receptors
nAChRs) based on acetylcholine (ACh) has led to the development of potent
agonists with clear preference for the α β nAChR, the most prevalent nAChR
(
4
2
subtype in the central nervous system. In this work we present the continuation of
these efforts aimed at increasing this subtype selectivity by introduction of
conformational restriction in the carbamoylcholine homologue, 3-
(dimethylaminobutyl) dimethylcarbamate (DMABC). Our results highlight the
importance of the N-carbamoyl substitution in α -subtype selectivity. Moreover,
4 2
β
we have confirmed the non-linear conformation of DMABC bound to nAChRs
suggested by recent crystal structures of the compound in complex with the
Lymnaea stagnalis ACh binding protein.
Keywords: Nicotinic acetylcholine receptor; Acetylcholine; Agonist;
Conformational restriction; Docking.
Abbreviations: CNS, central nervous system; PNS, peripheral nervous system;
ACh, acetylcholine; nAChR, nicotinic receptor; CCh, carbamoylcholine; Ls-
AChBP, Lymnasia stagnalis acetylcholine-binding protein.
2

ACCEPTED MANUSCRIPT
1
. Introduction
Acetylcholine (ACh) is a major neurotransmitter in the central nervous
system (CNS) and peripheral nervous system (PNS), where its actions are
mediated through two types of ACh receptors (AChRs): the metabotropic G
protein-coupled muscarinic AChRs (mAChRs) and a family of ligand-gated ion
channels, the nicotinic AChRs (nAChRs). Neuronal nAChRs are located at
presynaptic and postsynaptic neuron terminals through which they modulate
functions of several major neurotransmitter systems and influence numerous brain
functions [1–3]. For example, there is wide evidence supporting that nAChRs play
a role in a broad range of neurodegenerative and psychiatric disorders [4–8].
The nAChR complex consists of five subunits embedded in the cell
membrane that are symmetrically arranged to form a cation-selective channel pore.
In contrast to the muscle nAChRs expressed at the neuromuscular junction which
are α β γδ or α β γε assemblies, neuronal nAChRs are homomeric and heteromeric
1
1
1 1
2 7 9 2 4
assemblies of α (α -α , α -α10) and β (β -β ) subunits [9]. Among all the possible
receptor combinations, the vast majority of the nAChRs in the CNS are
heteromeric receptors comprising α and β subunits and homomeric α receptors,
4
2
7
whereas the most prevalent subtype in the PNS are α β receptors [10]. The
3
4
orthosteric binding site is situated in the interface between α and β subunits in the
heteromeric pentamer and between neighboring α subunits in the homomeric
3

ACCEPTED MANUSCRIPT
receptor [11–13]. Significant advances have been made in the understanding of the
complex molecular basis for nAChR signaling [14,15]. In order to investigate the
physiological and pathophysiological roles of nAChRs and potentially explore the
therapeutic potential in the receptors, the development of tools able to distinguish
between the different nAChR subtypes are needed [10,16]. However, the plethora
of physiologically expressed nAChR subtypes, and the highly conserved nature of
the orthosteric binding site in the receptors make the development of subtype-
selective agonists a challenging task [17].
In recent studies, we have developed a series of carbamoylcholine (CCh)
homologues that act as functionally α β -selective agonists (figure 1): 3-
4
2
(
dimethylamino)butyl methylcarbamate (2), 3-(dimethylamino)butyl azetidine-1-
carboxylate (3), [4-(1-methyl-1H-imidazol-2-yloxy)but-2-yl]-N,N-dimethylamine
4) and [4-(1,4-dimethyl-1H-imidazol-2-yloxy)but-2-yl]-N,N-dimethylamine (5).
(
CCh homologues 1 and 2 exhibit high degree of selectivity for nicotinic over
muscarinic receptors, and in addition 2-5 display a pronounced degree of
selectivity towards β -containing nAChRs subtypes [18–20]. Initially, these CCh
2
homologues were thought to bind to the nAChRs in a linear conformation similar
to that of ACh or CCh [15]. However, recent crystallographic studies of the
Lymnaea stagnalis ACh-binding protein (Ls-AChBP) in complex with 1 and 4
have found the compounds to bind Ls-AChBP in a folded conformation favored by
4

ACCEPTED MANUSCRIPT
an intramolecular hydrogen bond established between the protonated
dimethylamino group and the ester-like oxygen or the bioisosteric imidazole N3 of
these compounds (figure 1)[20].
In the observed binding mode in these co-crystal structures, the dimethyl
ammonium group of the agonist lies covered by the flexible C-loop of the α
subunit within an aromatic box, while the other end of the molecule faces the
complementary subunit, the main point of contact being a hydrogen bond from a
water molecule to the carbonyl oxygen of 1 (figure 2) and the N3 in the imidazole
ring of 4 (not shown) [20], respectively. This interaction is also present in the Ls-
AChBP co-crystal structure with ACh (not shown) and is considered a determinant
for agonist binding and function [15].
Besides the proposed hydrogen-bond interaction with a water molecule in
the interface with the complementary subunit, the carbamate/azole part of the
ligands is considered essential for the observed subtype-selectivity profile.
Plausible steric clashes against the complementary subunit are thought to be the
most relevant interaction for nAChR subtype selectivity [19,20]. Additionally, the
interaction with the complementary subunit has been suggested to be responsible
for the efficacy displayed by agonists at the α β nAChR [15,22].
4
2
5

ACCEPTED MANUSCRIPT
The amino part of 1 has already been studied by means of chain elongation,
resolution of the enantiomers, alpha methyl deletion or alkyl substitution [19], and
incorporation into cyclic structures [23,24]. Herein, we present a systematic
introduction of different cycles representing various degrees of constriction of the
conformational space of 1. A cyclopropyl moiety was included within the structure
of 1 and 2, aiming to lock the bio-active conformation of 1, including the putative
intramolecular hydrogen bond described above in the crystal structure of 1 in Ls-
AChBP. Similarly larger cycles showing different degrees of bulkiness and
flexibility were decorated with the main structural features of the parent
compounds giving a series of cyclic urea and thiourea analogues.
2
. Chemistry. Synthesis of restricted analogues of 1.
The synthetic route that leads to target compounds 13-16 is outlined in
scheme 1. Compound trans-8 was synthesized via selective enzymatic mono-
hydrolysis of 7 by pig liver esterase [25,26]. The hydrolysis of the inexpensive
cyclic anhydride 6 was chosen for the synthesis of cis-8 [27]. The key step in the
synthetic route was the Curtius rearrangement[28]; the diphenylphosphorylazide
mediated rearrangement worked with excellent yields for both trans-10 and cis-10.
6

ACCEPTED MANUSCRIPT
Unlike Csuk et al. reported previously, no isomerization of the cis isomer was
observed [29–31].
4 4
Different reducing agents (LiBH , LiAlH and DIBAL) were tested to
selectively reduce the carbonyl group of the ethylcarboxylate to the corresponding
alcohols (trans-10 and cis-10) of which DIBAL displayed the best performance
giving the best yields [26]. Subsequently the Boc-protected carbamates trans-11,
cis-11, trans-12 and cis-12 were synthesized in two steps: first, the
imidazolylcarboxylate was obtained from the corresponding alcohol by addition of
carbonyldiimidazole and then, the corresponding amine was added, to afford the
Boc-protected carbamates in good yields. The cleavage of the Boc group was
performed in a mixture of trifluoroacetic acid and dichloromethane [32]. The
yields of the primary amines trans-13, cis-13, trans-14 and cis-14 ranged from fair
to good. The primary amines underwent reductive alkylation with formaline and
sodium cyanoboronhydride to afford the dimethylated derivatives trans-15, cis-15,
trans-16 and cis-16 in excellent yields [33]. Both the primary amines and the
dimethylated ones were obtained as oxalate salt.
Compounds 24-31, 36-38 were synthesized by addition of a monomethyl or
Boc protected diamine to the corresponding isocyanate or isothiocyanate (scheme
2
). The reaction of the diamines with dimethylcarbamyl chloride yielded the
dimethylureas 18-20, 35. Amide 21 was synthesized via carbonyldiimidazole
7

ACCEPTED MANUSCRIPT
coupling followed by Boc-cleavage in HBr to afford compound 22. The same Boc-
cleaving procedure was employed to obtain compounds 23, 32 and 33.
3
. Pharmacology.
3
.1 Binding characteristics of selected compounds at α β , α β and α β nAChRs.
4
2
4
4
3 4
The binding properties of all synthesized compounds were determined in a
3 4
H]epibatidine competition binding assay to recombinant rat α , α and α β
3
[
4
β
2
4
β
4
nAChRs stably expressed in HEK293 cells. The results are given in table 1. In
contrast to 1 and 2, none of the synthetic cyclic homologues of 1 displayed
nanomolar binding affinity for the α β nAChR [18].
4
2
In the dimethylaminocyclopropyl series, the relative position of the
substituents introduced on the cycle was found to be a key determinant for
receptor binding, as the trans compounds (trans-15 and trans-16) were totally
devoid of affinity for any of the nAChRs tested. Conversely, the cis analogs (cis-
1
5 and cis-16) displayed a selectivity pattern similar to that of the reference
compounds 1 and 2, with the dimethylcarbamate cis-15 being slightly less
selective for α β over α β and α β than the monomethylcarbamate cis-16.
4
2
3
4
4 4
8

ACCEPTED MANUSCRIPT
All dimethylaminopyrrolidine analogs exhibited similar low micromolar
binding affinities. The dimethylurea derivative 35 showed no selectivity for any of
the tested nAChRs, whereas homologues with monosubstituted urea (36) and
thiorueas (37 and 38) exhibited increased degrees of α β selectivity. The
4
2
ethylthiourea 38 was the most selective ligand of the entire series
(α )/K (α ) ~250 fold, (K (α )/K (α ) >800 fold) exhibiting a selectivity
profile identical to that of 2 ((K (α β )/K (α β ) 220 fold, (K (α β )/K (α β ) 770
(K
i
4
β
4
i
4
β
2
i
3
β
4
i
4 2
β
i
4
4
i
4
2
i
3
4
i
4 2
fold). The (thio)urea substitution-selectivity pattern observed in this series of
compounds were similar to that observed for the carbamate-containing parent
compounds.
The urea and thiourea analogues comprising larger cycles than pyrrolidine
4 2
displayed quite weak binding affinities to the nAChRs. A preference for α β
receptors was observed in all analogs with monosubstituted ureas and thioureas
with significant binding affinities. Even if a full displacement curve could not be
obtained for the dimethylurea 18, this compound appeared to lack selectivity for
any subtype. Deletion of the N(amino)-methyl substituent, and replacement of the
2
3
sp nitrogen of 18 for a sp carbon gave the inactive compound 22. The
homopiperazine series (32 and 33) was totally inactive, the only exception being
the secondary amine 23, just as none of 1-methyl homopiperazine analogues (19,
2
5, 27, 29) were active at the receptors.
9

ACCEPTED MANUSCRIPT
3
.2 Functional properties of selected compounds at α β and the α β nAChRs.
4
2
3 4
Selected analogs were functionally characterized at α β and α β nAChRs
4
2
3 4
in the FLIPR Membrane Potential Blue assay (table 2)
.
All cycle-containing
compounds tested were found to be agonists, although we could not determine
their respective EC50 values due to their weak potencies, except in the case of 35.
This compound, as a racemic mixture, was as potent as other high affinity CCh
analogues like 3 (EC 7.1 µM) at the α β subtype but it did not display significant
5
0
4 2
selectivity over the α β nAChR. Potency of compound 35 at the α β is in the
3
4
3 4
same range as that of 1 (EC 7.7 µM) [19].
50
4
. Molecular modelling. Docking studies.
To explore putative interactions of the ligands with the nAChR,
representatives of the dimethylaminocyclopropyl series (cis-15 and trans-15) and
the dimethylaminopyrrolidine derivative 35, that originally were designed to lock
the folded bioactive conformation of (R)-1, were investigated in docking
experiments towards the binding site of Ls-AChBP co-crystalised with (R)-1 [20].
The
dimethylamino)cyclopropyl)methyl dimethyl carbamate; Glide XP score -12.1
kcal/mol) represented a low energy conformation (˂1 kcal/mol above the global
highest
scoring
pose
of
cis-15
(isomer
((1S,2R)-2-
(
10

ACCEPTED MANUSCRIPT
energy minimum) and revealed a putative binding mode closely resembling the co-
crystallised reference ligand (R)-1 (Figure 3A). The opposite diastereoisomer
(
1R,2S) revealed a comparable docking score (Glide XP score -12.0 kcal/mol) but
this docking mode was disregarded as it represented a high energy conformation
+3 kcal/mol above the global energy minimum). On the other hand, trans-15
(
scored worse (Glide XP scores -10.9 and 10.0 kcal/mol for (1S,2S) and (1R,2R)
configurations, respectively). Furthermore, the docked poses represented high
energy conformations (˃6 kcal/mol above the global energy minimum for both
diastereoisomers) in agreement with the observed lower affinity. Thus, when
locked in the trans configuration, it seems that this molecule cannot adopt a
conformation that can be reasonably superimposed on (R)-1. The dockings of
compound 35 resulted in similar scores (-12.6 and -9.1 kcal/mol for the S- and R-
enantiomers, respectively), and a binding mode for (S)-35 closely resembling that
of the reference compound was observed (Figure 3A). These data tentatively
suggest that the activity of 35 resides mostly in the S-enantiomer.
5
. Discussion
Considering the optimal overlap observed in docking between cis-15 and
5 and 1 the significantly lower binding affinities displayed by these two
3
analogues compared to the lead (75 and 320 fold relative to 1, respectively) are
1
1

ACCEPTED MANUSCRIPT
interesting. From previous studies, it is known that the C3-methyl group of 1 is of
key importance for the high binding affinity of these compounds to nAChRs [18].
Based on the docking results it seems plausible that cis-15 and 35 are unable to
occupy the space corresponding to that occupied by the C3-methyl group in 1, and
thus fail to form the same favorable interactions with the tyrosines on loop-C as 1
and other C3-methyl analogs are able to do. Furthermore, both cis-15 and 35
contain additional bulk compared to 1, and this bulk may result in steric clashes
with the receptor or result in suboptimal cation-π contacts within the aromatic box.
None of the compound series synthesized in this work exhibited a
selectivity profile attributable to the cycle used as scaffold. Conversely, all series
supported the hypothesis that the substitution of the N-carbamoyl moiety is the
determinant feature for selectivity in CCh analogues regardless of the aminoalkyl
scaffold. Based on the docking studies and available X-ray structures it is evident
that the substituents on carbamoyl moieties point towards the complementary
subunit which is a main determinant of subtype selectivity in nAChRs. In this area
of the protein the amino acid sequence varies between subunits [34]. We have
discussed this in detail elsewhere [19,20] and have recently demonstrated how
mutation of residues in this area controls subtype specificity of ligands [35].
Structural observations suggested that the interactions with the complementary
subunit that putatively permit compound 2 to discriminate between different
12

ACCEPTED MANUSCRIPT
nAChR subtypes better than 1 are reproduced in the case of the cis-15 - cis-16
pair. The same applies in the case of disubstituted urea 35 compared to the
monosubstituted ureas and thioureas 36-38 in the pyrrolidine series. The
agreement found in the structure-selectivity relationship further supports that the
binding to nAChRs of these cyclic analogues should occur in a similar manner to
that of dimethylaminobutyl derivatives 1 and 2.
The structure-function relationship that determines the intrinsic activity of
the orthosteric nAChRs ligand remains elusive. Numerous efforts have been made
in order to rationalize the mechanistic determinants underlying agonist activity..
Using Ls-AChBP as α β surrogate, the closure of the C-loop in the α subunit has
4
2
been proposed to correlate with agonist efficacy [36,37]. More recently, it has
been proposed that C-loop closure mostly determines binding affinity to the
receptor, whereas ligands efficacy relies on their ability to stabilize a strong
intersubunit bridge between the interfaces of the primary α subunit and the
complementary subunit [22,38]. As outlined in the Results section, all cycle-
containing analogs characterized functionally in this study were found to be
agonists at the α β subtype regardless of their cyclic scaffold, and this agonism is
4
2
more likely to be attributable to the different substituted urea and thiourea ring
systems than to the amino-alkylic core of the compounds. The comparative
biostructural analysis of all four possible diastereomers of compound 15 docked
1
3

ACCEPTED MANUSCRIPT
into Ls-AChBP supports the preferential affinity of α β nAChR for the cis-15
4
2
racemate compared to trans-15. The correspondence between the preferred relative
configuration and displayed affinity highlights the clear preference of α β nAChR
4
2
for structures resembling that of folded-1. This statement is further supported in
the case of cyclopropyl-urea 35 thus confirming that binding of compound 1 to
4 2
nAChR α β should ocurr in a folded manner.
6. Conclusions
A series of cyclic DMABC (1) analogues have been synthesized and
characterized pharmacologically at different nAChRs subtypes. The overall
conclusion emerging from this work is that the size of cycle introduced in the
analogs correlated inversely with their nAChR binding affinities, with
aminocyclopropane and aminopyrrolidine analogs exhibiting the highest binding
affinities. The results support the notion of the importance of N-carbamate
substitution for the selectivity SAR displayed by the CCh homolog derivatives,
and demonstrate that substituted urea and thiourea rings can partially substitute for
this moiety. In line with this, racemic compound 35 has proved to be functionally
as potent as some ligands from the flexible homologues series. Although the
conformational restriction approach used in this work has failed to provide potent
selective α β nAChR ligands, the comparative structural analysis of compound 35
4
2
14

ACCEPTED MANUSCRIPT
and cis-15 enantiomers strongly supports the folded conformation of compound 1
in its binding to the α β nAChR. Having confirmed the folded conformation as a
4
2
likely bioactive conformation, the study provides a solid foundation for future
structure based projects in which potency of compounds can be improved by
inspiration from other agonist/AChBP structures. Combined with studies of the
individual enantiomers, this may lead to new potent agonists
7
. Experimental section
7
7
.1. Chemistry.
.1.1. General procedures. All reactions involving air sensitive reagents were
2
performed under a N atmosphere using syringe-septum cap techniques and
glassware was flame-dried in vacuo prior to use. Unless otherwise noted, all
materials were obtained from commercial suppliers and used without further
purification. When dry solvents were used, THF was distilled and stored over
molecular sieves (4Å) and toluene was purchased from a commercial source and
1
13
stored over molecular sieves (4Å). H NMR and C NMR spectra were obtained
on a Varian Gemini 2000 (300 MHz) or Varian Mercury Plus (300 MHz).
Analytical thin-layer chromatography (TLC) was carried out using Merck silica
gel 60 F254 plates, and the compounds were visualized using either KMnO
4
spraying reagent or phosphomolybdic acid 10% wt. in ethanol. Elemental analyses
were performed by J. Theiner at Department of Physical Chemistry, University of
1
5

ACCEPTED MANUSCRIPT
Vienna, Austria. Melting points were determined in open capillary tubes and are
uncorrected. Liquid chromatography was performed on a Waters analytical LC-
MS equipped with a SunFire C18 4.6 x 50 mm column and a cuadrupole mass
spectrometer (Micromass ZQ).
Oxalate salt formation procedure: To the free base of the amine (1 equiv.) a
solution of oxalic acid (1.0 – 1.3 equiv.) dissolved in warm acetone was added,
and the mixture was cooled to force crystallization. In the cases were this was not
sufficient, after the evaporation of the solvent, the mixture was recrystallized from
EtOH:Et O.
2
7
.1.1.1. General method (A) for N,N-dimethylsubsituted ureas: The
corresponding amine (1 eq.) was added carefully over a solution of
dimethylcarbamyl chloride (1 eq.) and trimethylamine (2 eq.) in anhydrous THF (5
mL per mmol). The evolution was followed by TLC. Upon completion the formed
precipitate (triethylammonium chloride) was filtered off, and the filtrate
evaporated. When necessary, the residue was resuspended in water and basified
with NaOH 2M to pH 11-12. The aqueous suspension was subjected to liquid-
liquid continuous extraction with EtOAc that, after drying over MgSO , and
4
evaporation of the solvent gave the free base.
16

ACCEPTED MANUSCRIPT
7
.1.1.2. General method (B) for Boc-cleavage: The corresponding Boc-protected
amine was dissolved in EtOH (4 mL per mmol) and a solution of HBr 62% aq.
0.3 mL per mmol) was added. The mixture was kept stirring overnight, then the
solvent removed and the resulting hydrobromide salt recrystalised from
EtOH:Et O.
(
2
7
.1.1.2. General method (C) for N-monosubsituted ureas and thioureas: The
amine (1 eq.) was dissolved in Et O (4 mL per mmol), and the corresponding
2
isocyanate or isothyocyanate (1.2 eq.) was then added. The evolution of the
reaction was followed by TLC. Generally completion was achieved after 0.5 h
stirring at room temperature. The corresponding product was either filtered off as a
precipitate or recovered after evaporation of the solvent.
7
.1.2. (±)-cis -ethyl 2-((tert-butoxycarbonyl)amino)cyclopropanecarboxylate
((±)-cis-9).
To a stirred solution of (±)-cis-8 (4.3 g, 27.4 mmol) in dry toluene (30 mL), dry
triethylamine (2.7 g, 27.4mmol) and dry tert-butanol (25.7 g, 347 mmol) diphenyl-
phosphorylazide (7.52 g, 27.4 mmol) was added. The solution was heated to 55ºC
under nitrogen and stirred for 24 h. After cooling to room temperature the solvent
and the excess of tert-butanol were removed under reduced pressure, the
1
7

ACCEPTED MANUSCRIPT
remaining oil resuspended in EtOAc (40 mL) and washed with citric acid 10% (2 x
3
0 mL). The acidic washings were extracted with EtOAc (20 mL). The combined
organics were then washed with NaHCO sat. (30 mL) and brine (30 mL), dried
3
(
MgSO ) and solvent evaporated to afford (±)-cis-9 (5.71 g, 91%) as a white
4
1
crystals. Mp: 53-55ºC. H NMR (300 MHz, CDCl
3
) δ 5.55 – 5.08 (s, 1H), 4.30 –
3
1
1
1
.98 (qd, J = 7.1, 1.1 Hz, 2H), 3.45 – 3.20 (m, 1H), 1.93 – 1.76 (q, J = 7.7 Hz,
H), 1.46 – 1.39 (s, 9H), 1.31 – 1.24 (t, J = 7.1 Hz, 4H), 1.19 – 1.11 (q, J = 5.9 Hz,
1
3
H). C NMR (75 MHz, CDCl ) δ 172.12, 156.19, 79.85, 61.05, 31.70, 28.70,
3
-1
9.36, 14.57, 14.31. IR ( cm ): 3361, 2980, 2935, 1710, 1512, 1183, 1155.
7
.1.3.
(±)-trans-(2-((tert-butoxycarbonyl)amino)cyclopropyl)methyl
dimethylcarbamate ((±)-trans-11).
To a solution of (±)-trans-10 (704 mg, 3.76 mmol) in a mixture anhydrous toluene
(11 mL) and anhydrous THF (3 mL) carbonyldiimizadole (914 mg, 5.64 mmol)
and a catalytic amount of 4-dimethylaminopyridine were added. After stirring at
room temperature for 2.5 h a solution of dimethylamine 33% in EtOH was added
(
18.8 mL) and stirred for an additional hour. The reaction was monitored on TLC.
The solvent was eliminated under vacuum and the resulting oil dissolved in EtOAc
30 mL). The solution was washed with citric acid 10% in H O until the pH stayed
(
2
4
acidic after washing, then NaOH 1 M (2 x 10 mL), brine (10mL), dried (MgSO )
18

ACCEPTED MANUSCRIPT
1
and solvent evaporated to afford (±)-trans-11 (800 mg, 83%) as a colorless oil. H
NMR (300 MHz, CDCl ) δ 4.88 – 4.75 (br s, 1H), 3.99 – 3.90 (m, 2H), 2.87 (d, J =
3
0
8
3
1
.9 Hz, 6H), 2.48 – 2.41 (m, 1H), 1.40 (s, 9H), 1.31 – 1.16 (m, 1H), 0.73 (dd, J =
1
3
.1, 5.2 Hz, 2H). C NMR (75 MHz, CDCl ) δ 156.61, 156.34, 79.61, 66.95,
3
-1
6.61, 36.13, 28.60, 28.33, 19.81, 12.37. IR (cm ): 3332, 2976, 2932, 1686, 1498,
163.
7
.1.4.
(±)-cis-(2-((tert-butoxycarbonyl)amino)cyclopropyl)methyl
dimethylcarbamate ((±)-cis-11).
According to the preparation of (±)-trans-11, from (±)-cis-10 (1.12 g, 6.01
1
mmol) (±)-cis-11 (1.19 g, 76%) was obtained as white crystal. Mp: 84.5-87.0ºC. H
NMR (300 MHz, CDCl
3
) δ 5.72 – 5.67 (br s, 1H), 4.26 (dd, J = 11.9, 4.6 Hz, 1H),
3
1
.96 (dd, J = 11.9, 9.4 Hz, 1H), 2.91 (s, 6H), 2.70 – 2.57 (m, 1H), 1.43 (s, 9H),
.30 – 1.17 (m, 1H), 1.00 (dt, J = 12.4, 5.8 Hz, 1H), 0.39 (td, J = 6.1, 4.2 Hz, 1H).
1
3
C NMR (75 MHz, CDCl ) δ 157.52, 157.13, 79.48, 64.40, 36.80, 36.28, 28.67,
3
-1
2
7.58, 17.01, 11.02. IR(cm ): 3326, 2977, 2933, 1685, 1500, 1387, 1365, 1165.
7
.1.5
(±)-trans-tert-butyl
(2-
(((methylcarbamoyl)oxy)methyl)cyclopropyl)carbamate ((±)-trans-12).
1
9

ACCEPTED MANUSCRIPT
According to the preparation of (±)-trans-11, from (±)-trans-10 (1.45 g, 7.75
mmol), employing methylamine 33% in EtOH instead (3.6 mL), (±)-trans-12
1
(
1.48 g, 78%) was obtained as a white solid. Mp: 94.7-95.2ºC. H NMR (300
MHz, CDCl ) δ 4.67 (br s, 2H), 3.95 (d, J = 8.3 Hz, 2H), 2.79 (d, J = 4.2 Hz, 3H),
3
2
0
2
.50 – 2.45 (m, 1H), 1.68 – 1.62 (m, 1H), 1.45 (s, 9H), 1.30 – 1.22 (m, 1H), 0.82 –
1
3
.72 (m, 1H). C NMR (75 MHz, CDCl
3
) δ 157.24, 156.37, 79.81, 66.72, 28.66,
-1
8.47, 27.82, 19.87, 12.43. IR (cm ): 3338, 2977, 1693, 1524, 1253, 1167.
7.1.6.
(±)-cis-tert-butyl(2-
(((methylcarbamoyl)oxy)methyl)cyclopropyl)carbamate ((±)-cis-12).
According to the preparation of (±)-trans-11, from (±)-cis-10 (885 mg, 4.73
mmol), employing methylamine 33% in EtOH (3.4 mL) (±)-cis-12 (866 mg, 75%)
-1
was obtained as white crystal. Mp: 92.8-95.2ºC. IR(cm ): 3327, 2977, 1689, 1514,
1
1
253, 1164. H NMR (300 MHz, CDCl
3
) δ 5.60 – 5.49 (br s, 1H), 4.81 – 4.72 (br
s, 1H), 4.24 (dd, J = 11.8, 4.9 Hz, 1H), 3.95 (t, J = 10.6 Hz, 1H), 2.79 (d, J = 5.4
Hz, 3H), 2.67 – 2.59 (m, 1H), 1.44 (s, 9H), 1.30 – 1.17 (m, 1H), 1.04 – 0.96 (m,
1
3
1
7
H), 0.41 (td, J = 6.1, 4.2 Hz, 1H). C NMR (75 MHz, CDCl ) δ 157.50, 157.09,
3
9.59, 64.13, 28.70, 27.84, 27.53, 16.89, 11.16.
20

ACCEPTED MANUSCRIPT
7
1
.1.7. (±)-trans-(2-aminocyclopropyl)methyl dimethylcarbamate ((±)-trans-
3).
A solution of (±)-trans-11 (688 mg, 2.68 mmol) in CH Cl (10 mL) and TFA (10
2
2
mL) was stirred at room temperature for 30 min after which solvent was
evaporated to afford the a crude. The reaction crude was dissolved in H O (20 mL)
2
and basified with conc. NaOH. The free amine was extracted from the solution
through continuous liquid-liquid extraction with CH Cl . The solvent was dried
2
2
1
(
MgSO ) and evaporated to afford (±)-trans-13 (326 mg, 77%) as a yellow oil. H
4
NMR (300 MHz, CDCl ) δ 3.87 (qd, J = 11.5, 7.2 Hz, 2H), 2.90 (s, 6H), 2.70 (s,
3
2
0
3
H), 2.34 – 2.29 (m, 1H), 1.27 – 1.12 (m, 1H), 0.66 (ddd, J = 9.2, 5.3, 4.0 Hz, 1H),
1
3
.54 (dt, J = 7.1, 5.4 Hz, 1H). C NMR (75 MHz, CDCl ) δ 156.83, 67.75, 36.66,
3
-1
6.19, 30.04, 20.58, 12.80. IR(cm ): 3393, 2941, 1698,1459, 1390, 1185.
7
.1.8. (±)-cis-(2-aminocyclopropyl)methyl dimethylcarbamate ((±)-cis-13).
According to the preparation of (±)-trans-13, from (±)-cis-11 (1.15 g, 4.5 mmol)
1
(
±)-cis-13 (460 mg, 65%) was obtained as an orange oil. H NMR (300 MHz,
CDCl ) δ 4.39 (dd, J = 11.4, 5.5 Hz, 1H), 4.10 (dd, J = 11.3, 9.2 Hz, 1H), 2.92 (s,
3
6
0
H), 2.50 (td, J = 6.9, 4.1 Hz, 1H), 1.73 (s, 2H), 1.09 (ddd, 12.5, 6.3, 3.4 Hz, 1H),
1
3
.77 (ddd, J = 9.1, 6.8, 5.2 Hz, 1H), 0.27 (dt, J = 5.7, 4.2 Hz, 1H). C NMR (75
2
1

ACCEPTED MANUSCRIPT
-1
MHz, CDCl ) δ 156.99, 64.82, 36.64, 36.24, 28.59, 17.31, 12.32. IR (cm ): 3378,
3
2
937, 1686, 1408, 1187.
7.1.9. (±)-trans-(2-aminocyclopropyl)methyl methylcarbamate ((±)-trans-14).
According to the preparation of (±)-trans-13, from (±)-trans-12 (1.32 g, 5.4
1
mmol) (±)-trans-14 (432 mg, 56%) was obtained as a colourless oil. H NMR (300
MHz, CDCl ) δ 4.72 – 4.56 (br s, 1H), 3.95 – 3.77 (m, 2H), 2.79 (d, J = 4.9 Hz,
3
3
H), 2.28 (dt, J = 6.9, 3.4 Hz, 1H), 1.72 (s, 2H), 1.16 – 1.06 (m, 1H), 0.61 (ddd, J
1
3
=
9.0, 5.1, 3.9 Hz, 1H), 0.50 (dt, J = 6.9, 5.3 Hz, 1H). C NMR (75 MHz, CDCl )
3
-1
δ 157.83, 66.75, 29.05, 27.82, 19.95, 12.24. IR (cm ): 3350, 2956, 1693, 1536,
1
265, 1139.
7.1.10. (±)-cis-(2-aminocyclopropyl)methyl methylcarbamate ((±)-cis-14).
According to the preparation of (±)-trans-13, from (±)-cis-12 (826 mg, 3.38
1
mmol) (±)-cis-14 (340 mg, 70%) was obtained as a colourless oil. H NMR (300
MHz, CDCl ) δ 4.72 – 4.63 (br s, 1H), 4.39 (dd, J = 11.3, 5.5 Hz, 1H), 4.08 (dd, J
3
=
11.3, 9.1 Hz, 1H), 2.79 (d, J = 4.9 Hz, 3H), 2.49 (td, J = 6.8, 4.1 Hz, 1H), 1.72
(s, 2H), 1.17 – 0.99 (m, 1H), 0.75 (ddd, J = 9.0, 6.8, 5.2 Hz, 1H), 0.24 (td, J = 5.6,
1
3
4
1
.1 Hz, 1H). C NMR (75 MHz, CDCl ) δ 157.51, 64.42, 28.55, 27.85, 17.22,
3
-1
2.25. IR (cm ): 3341, 2924, 1692, 1535, 1259, 1135.
22

ACCEPTED MANUSCRIPT
7
.1.11. (±)-trans-(2-(dimethylamino)cyclopropyl)methyl dimethylcarbamate
((±)-trans-15).
To a solution of (±)-trans-13 (133 mg, 0.85 mmol) in H O (4 mL) MeCN (5 mL)
2
2 3
formaldehyde 37% in H O (670 µL) NaCNBH (165 mg, 2.66 mmol) was added.
The pH was adjusted and maintained at 6 by addition of glacial AcOH and the
mixture stirred at room temperature for 1 h. The reaction mixture was quenched
with few drops of conc. HCl and washed with EtOAc (2 x 10 mL). The pH was
then turned basic by addition of conc. NaOH and extracted with CH Cl (4 x 10
2
2
mL). The organic layer was dried (MgSO ) and solvent evaporated to afford (±)-
4
1
trans-15 (127 mg, 65%) as a colourless oil. H NMR (300 MHz, CDCl ) δ 4.09
3
(dd, J = 11.4, 6.1 Hz, 1H), 3.67 (dd, J = 11.4, 8.7 Hz, 1H), 2.92 (s, 6H), 2.34 (s,
6
H), 1.57 (dt, J = 6.9, 3.5 Hz, 1H), 1.21 (ddd, J = 8.8, 5.9, 3.0 Hz, 1H), 0.74 (dt, J
1
3
=
3
9.2, 4.5 Hz, 1H), 0.54 (dt, J = 6.8, 5.3 Hz, 1H). C NMR (75 MHz, CDCl ) δ
-
1
1
1
56.75, 67.76, 45.87, 45.15, 36.65, 36.13, 19.85, 12.21. IR (cm ): 3393, 2940,
908, 1459, 1390, 1185. The oxalate salt was formed according to the general
method and thus from (±)-trans-15 (103 mg, 0.55 mmol) (±)-trans-15 oxalate
1
(
101 mg, 66%) was obtained as white crystals. Mp: 109.6 -110.5ºC. H NMR (300
MHz, D O) δ 4.03 (dd, J = 11.6, 5.8 Hz, 1H), 3.59 (dd, J = 11.6, 8.3 Hz, 1H), 2.80
2
(s, 6H), 2.76 – 2.70 (m, 6H; dd, J = 7.6, 3.5 Hz, 1H), 1.62 (ttd, J = 11.7, 6.2, 3.5
2
3

ACCEPTED MANUSCRIPT
Hz, 1H), 1.10 (ddd, J = 10.5, 7.3, 4.3 Hz, 1H), 0.94 (dd, J = 7.3, 7.5 Hz, 1H).
+
HPLCMS (m/z) [MH ] 187.32 (free base). Anal. calcd for C H N O · 0.95
9
18
2
2
C H O : C, 47.82; H, 7.30; N, 10.14. Found: C, 48.34; H, 7.22; N, 10.08.
2
2
4
7.1.12. (±)-cis-(2-(dimethylamino)cyclopropyl)methyl dimethylcarbamate
oxalate ((±)-cis-15 oxalate).
According to the preparation of (±)-trans-15, from (±)-cis-13 (293 mg, 1.85
1
mmol) (±)-cis-15 (325 mg, 94%) was obtained as a bright yellow oil. H NMR
(
300 MHz, CDCl ) δ 4.36 (dd, J = 11.2, 6.6 Hz, 1H), 4.02 (dd, J = 11.2, 8.8 Hz,
3
1
7
1
1
H), 2.92 (s, 6H), 2.33 (s, 6H), 1.73 (td, J = 6.8, 4.4 Hz, 1H), 1.20 (tdd, J = 12.6,
.5, 4.6 Hz, 1H), 0.73 (ddd, J = 9.0, 6.8, 5.2 Hz, 1H), 0.32 (dt, J = 5.7, 4.7 Hz,
1
3
H). C NMR (75 MHz, CDCl
3
) δ 156.39, 65.20, 48.96, 45.87, 44.90, 18.24,
-1
0.26. IR (cm ): 3372, 2942, 1681, 1457, 1190, 1056. The oxalate salt was
formed according to the general method and thus from (±)-cis-15 (320 mg, 1.72
mmol) (±)-cis-15 oxalate (233 mg, 84%) was obtained as white crystals. Mp:
1
8
4.7-85.6. H NMR (300 MHz, D O) δ 4.48 (dd, J = 12.6, 5.6 Hz, 1H), 3.69 (dd, J
2
=
12.4, 10.6 Hz, 1H), 2.86 (s, 6H), 2.82 – 2.70 (m, 1H; m 6H), 1.59 (dtd, J = 13.2,
9
.9, 6.9 Hz, 1H), 1.12 (qd, J = 7.5, 3.8 Hz, 1H), 0.73 (td, J = 7.1, 5.0 Hz, 1H).
+
HPLCMS (m/z) [MH ] 187.39 (free base). Anal. calcd for C H N O · 1.55
9
18
2
2
2 2 4
C H O : C, 45.8; H, 6.92; N, 10.68. Found: C, 42.46; H, 6.85; N, 10.93.
24

ACCEPTED MANUSCRIPT
7
.1.13. (±)-trans-(2-(dimethylamino)cyclopropyl)methyl methylcarbamate
oxalate ((±)-trans-16 oxalate).
According to the preparation of (±)-trans-15, from (±)-trans-14 (250 mg, 3.38
1
mmol) (±)-trans-16 (284 mg, 95%) was obtained as a colourless oil. H NMR (300
3
MHz, CDCl ) δ 4.72 – 4.54 (br s, 1H), 3.97 (dd, J = 11.5, 6.6 Hz, 1H), 3.78 (dd, J
=
11.3, 8.1 Hz, 1H), 2.80 (d, J = 4.9 Hz, 3H), 2.37 (s, 6H), 1.62 (dt, J = 6.7, 3.3
1
3
Hz, 1H), 1.32 – 1.16 (m, 1H), 0.83 – 0.74 (m, 1H), 0.55 (q, J = 5.7 Hz, 1H). C
NMR (75 MHz, CDCl ) δ 156.84, 67.23, 45.88, 45.15, 27.86, 19.72, 12.01. IR
3
-1
(
cm ): 3336, 2949, 1694, 1534, 1421, 1263, 1059. The oxalate salt was formed
according to the general method and thus from (±)-trans-16 (299 mg, 1.74 mmol)
(±)-trans-16 oxalate (324 mg, 71%) was obtained as white crystals. Mp: 125.4-
1
1
1
1
26.4ºC. H NMR (300 MHz, D
2
O) δ 3.95 (dd, J = 11.7, 6.0 Hz, 1H), 3.59 (dd, J =
1.8, 8.1 Hz, 1H), 2.79 – 2.78 (s, 6H; m, 1H), 2.52 (s, 3H), 1.66 – 1.49 (m, 1H),
+
.14 – 1.00 (m, 1H), 0.98 – 0.82 (m, 1H). HPLCMS (m/z) [MH ] 173.29 (free
base). Anal. calcd for C H N O · 1.1 C H O : C, 45.80; H, 6.92; N, 10.68.
8
16
2
2
2
2
4
Found: C, 44.28; H, 6.62; N, 11.13.
2
5

ACCEPTED MANUSCRIPT
7
.1.14.
(±)-cis-(2-(dimethylamino)cyclopropyl)methyl
methylcarbamate
oxalate ((±)-cis-16 oxalate).
According to the preparation of (±)-trans-15, from (±)-cis-14 (199 mg, 1.38
1
mmol) (±)-cis-16 (211 mg, 90%) was obtained as a yellow oil. H NMR (300
MHz, CDCl
3
) δ 4.67 (s, 1H), 4.27 (dd, J = 11.0, 6.3 Hz, 1H), 4.09 (dd, J = 11.1,
.3 Hz, 1H), 2.79 (d, J = 4.9 Hz, 3H), 2.29 (s, 6H), 1.69 (td, J = 6.8, 4.3 Hz, 1H),
.21 – 1.09 (m, 1H), 0.71 (ddd, J = 8.9, 6.8, 5.1 Hz, 1H), 0.27 (dt, J = 6.0, 4.6 Hz,
8
1
1
1
1
3
H). C NMR (75 MHz, CDCl ) δ 157.57, 64.86, 45.91, 44.71, 27.84, 18.14,
3
-1
0.50. IR (cm ): 3339, 2945, 1694, 1534, 1459, 1260, 1139. The oxalate salt was
formed according to the general method and thus from (±)-cis-16 (210 mg, 1.28
mmol) (±)-cis-16 oxalate (187 mg, 56%) was obtained as white crystals. Mp:
1
1
03.5-104.8ºC. H NMR (300 MHz, D
2
O) δ 4.49 (dd, J = 12.6, 5.4 Hz, 1H), 3.70
(dd, J = 12.5, 10.0 Hz, 1H), 2.90 – 2.83 (s, 6H; m, 1H), 2.52 (s, 3H), 1.63 – 1.48
(m, 1H), 1.11 (dt, J = 9.3, 7.3 Hz, 1H), 0.73 (td, J = 7.2, 4.9 Hz, 1H). HPLCMS
+
(
m/z) [MH ] 173.17 (free base). Anal. calcd for C H N O · 1.15 C H O : C,
8
16
2
2
2
2
4
4
5.8; H, 6.92; N, 10.68. Found: C, 43.71; H, 6.72; N, 11.15.
7.1.15. N,N,4-trimethylpiperazine-1-carboxamide oxalate (18)
Following method A, starting from 1-methylhomopiperazine (17a, 221 mg, 2.21
mmol) the free base 18 (245 mg, 65%) was obtained as a slightly yellowish oil.
26

ACCEPTED MANUSCRIPT
Spectroscopic data was in agreement with previous report [39]. The free base was
crystalised as an oxalate according to the general method and thus (18) was
1
obtained as a white solid. Mp: 112-114ºC. H NMR (400 MHz, D O) δ 3.83 (d, J =
2
1
6
3.2 Hz, 2H), 3.55 (d, J = 11.0 Hz, 2H), 3.27 – 3.11 (m, 4H), 2.95 (s, 3H), 2.90 (s,
1
3
H). C NMR (101 MHz, CDCl
3
) δ 167.14, 55.54, 46.17, 45.50, 40.20. HPLCMS
O·1.3 C : C, 45.97;
+
(m/z) [MH ] 172.1 (free base). Anal. Calcd for C
8
H
17
N
3
2 2 4
H O
H, 7.33; N, 16.08. Found: C, 44.05; H, 6.72; N, 14.38.
7
.1.16. N,N,4-trimethyl-1,4-diazepane-1-carboxamide oxalate (19).
Following method A, starting from 1-methylhomopiperazine (17b, 180 mg, 1.6
1
mmol), the free base 19 (200 mg, 67%) was obtained as a yellowish oil. H NMR
3
(400 MHz, CDCl ) δ 3.42 (t, J = 4.8 Hz, 1H), 3.37 (t, J = 6.1 Hz, 1H), 2.71 (s,
6
H), 2.67 (t, J = 4.8 Hz, 1H), 2.57 (t, J = 5.2 Hz, 1H), 2.33 (s, 1H), 1.90 (q, J = 6.0
1
3
Hz, 1H). C NMR (101 MHz, CDCl
3
) δ 165.47, 58.17, 57.86, 48.30, 48.10, 46.66,
3
9.06, 28.04. The free base was converted to the oxalate salt according to the
1
general procedure to obtain white crystals. Mp: 84-87ºC. H NMR (400 MHz,
D O) δ 3.75 (dd, J = 16.0, 5.2 Hz, 1H), 3.69 – 3.34 (m, 5H), 3.27 (t, J = 12.0 Hz,
2
+
1
H), 2.96 (s, 3H), 2.86 (s, 6H), 2.33 – 2.05 (m, 3H). HPLCMS (m/z) [MH ] 186.2
(
free base). Anal. calcd for C H N O·1.6 C H O : C, 44.5; H, 6.79; N, 12.76.
9 19 3 2 2 4
Found: C, 44.33; H, 6.95; N, 12.85.
2
7

ACCEPTED MANUSCRIPT
7.1.17. tert-butyl 4-(dimethylcarbamoyl)-1,4-diazepane-1-carboxylate (20).
Following method A, starting from 1-tert-butoxycarbonyl-homopiperazine (17c,
1
2
00 mg, 1 mmol), 20 (206 mg, 100%) was obtained as a colourless oil. H NMR
(400 MHz, CDCl
3
) δ 3.55 – 3.39 (m, 6H), 3.34 (t, J = 5.8 Hz, 2H), 2.81 (s, 6H),
1
.90 (m, 2H), 1.46 (s, 9H).
7.1.18. tert-butyl 4-(dimethylcarbamoyl)piperidine-1-carboxylate (21).
1
-(tert-butoxycarbonyl) piperidine-4-carboxylic acid (229 mg, 1 mmol) was
dissolved in anhydrous THF (4 mL), 1,1’-carbonyldiimidazole (190 mg, 1.10
mmol) and triethylamine (154 µL, 1.10 mmol) were added to the solution. After
0
.5 h at 50ºC a solution of dimethylamine 2M in THF (802 µL) were added. The
reaction was kept at room temperature overnight. In order to remove the imidazole
NaOH 2M was added to the mixture and then extracted with EtOAc , the organic
extracts were pooled together, dried over MgSO , and evaporated to give 21 (128
4
mg, 77%) as a clear oil. Spectroscopy data was in agreement with previous report
[40]
7.1.19. N,N-dimethylpiperidine-4-carboxamide hydrobromide (22).[41]
28