Structural isomers of saligenin-based β2-agonists: Synthesis and insight into the reaction mechanism

Article abstract of DOI:10.1039/d0ob02095h

Salmeterol and albuterol are well-known β2-adenoreceptor agonists widely used in the treatment of inflammatory respiratory diseases, such as bronchial asthma and chronic obstructive pulmonary disease. Here we report the preparation of structural isomers of salmeterol and albuterol, which can be obtained from the same starting material as the corresponding β2-agonists, depending on the synthetic approach employed. Using 1D and various 2D NMR measurements, we determined that the structure of prepared isomers holds the β-aryl-β-aminoethanol moiety, in contrast to the α-aryl-β-aminoethanol moiety found in salmeterol and albuterol. We investigated the reaction of β-halohydrin and amines responsible for the formation of β-aryl-β-amino alcohol-both experimentally and using computational methods. The structure of β-halohydrin with the methyl salicylate moiety imposes the course of the reaction. The solvent plays a relevant, yet ambiguous role in the direction of the reaction, while the strength of the base influences the reaction yield and isomer ratio in a more evident way. Using computational methods, we have shown that the most probable reaction intermediate responsible for the formation of the unexpected isomer is the corresponding para-quinone methide, which can be formed due to phenol present in the methyl salicylate moiety. After successful preparation of albuterol and salmeterol isomers, we tested their inhibition potency to human acetylcholinesterase (AChE) and usual and atypical butyrylcholinesterase (BChE). Kinetic studies revealed that both isomers are low-potency reversible inhibitors of human cholinesterases.

Full text of DOI:10.1039/d0ob02095h

Organic &
Biomolecular Chemistry
View Article Online
View Journal
PAPER
Structural isomers of saligenin-based
β2-agonists: synthesis and insight into the
Cite this: DOI: 10.1039/d0ob02095h
reaction mechanism†
a,b
c,d
e
f
Anamarija Knežević,
*
Jurica Novak, Anita Bosak and Marijana Vinković
Salmeterol and albuterol are well-known β2-adenoreceptor agonists widely used in the treatment of
inflammatory respiratory diseases, such as bronchial asthma and chronic obstructive pulmonary disease.
Here we report the preparation of structural isomers of salmeterol and albuterol, which can be obtained
from the same starting material as the corresponding β2-agonists, depending on the synthetic approach
employed. Using 1D and various 2D NMR measurements, we determined that the structure of prepared
isomers holds the β-aryl-β-aminoethanol moiety, in contrast to the α-aryl-β-aminoethanol moiety found
in salmeterol and albuterol. We investigated the reaction of β-halohydrin and amines responsible for the
formation of β-aryl-β-amino alcohol – both experimentally and using computational methods. The struc-
ture of β-halohydrin with the methyl salicylate moiety imposes the course of the reaction. The solvent
plays a relevant, yet ambiguous role in the direction of the reaction, while the strength of the base influ-
ences the reaction yield and isomer ratio in a more evident way. Using computational methods, we have
shown that the most probable reaction intermediate responsible for the formation of the unexpected
isomer is the corresponding para-quinone methide, which can be formed due to phenol present in the
methyl salicylate moiety. After successful preparation of albuterol and salmeterol isomers, we tested their
inhibition potency to human acetylcholinesterase (AChE) and usual and atypical butyrylcholinesterase
Received 15th October 2020,
Accepted 16th November 2020
DOI: 10.1039/d0ob02095h
(BChE). Kinetic studies revealed that both isomers are low-potency reversible inhibitors of human
rsc.li/obc
cholinesterases.
1. Introduction
The vicinal amino alcohol moiety is a commonly found motif
in naturally occurring molecules, from hydroxy amino acids to
a
Division of Organic Chemistry and Biochemistry, Ruđer Bošković Institute, Bijenička
cesta 54, 10000 Zagreb, Croatia. E-mail: Anamarija.Knezevic@irb.hr
1
,2
b
lipids and sugars. Furthermore, it appears frequently in the
structures of many synthetic molecules, from biologically
Faculty of Science and Technology, University of Twente, Drienerlolaan 5, 7522 NB
Enschede, The Netherlands
c
3
Laboratory of Computational Modeling of Drugs, Higher Medical and Biological
active molecules to various catalysts. In fact, the β-amino
School, South Ural State University, 20-A, Tchaikovsky Str., Chelyabinsk 454080,
Russia
alcohol fragment is the key element in numerous pharmaco-
logically important molecules such as antibiotics, β-blockers,
β-adrenergic agonists, etc. The latter, potent bronchodilators
widely used in the treatment of bronchial asthma, are in struc-
tural correlation with norepinephrine, a physiological neuro-
d
Division of Physical Chemistry, Ruđer Bošković Institute, Bijenička cesta 54, Zagreb
0000, Croatia
1
e
Biochemistry and Organic Analytical Chemistry Unit, Institute for Medical Research
and Occupational Health, Ksaverska cesta 2, 10000 Zagreb, Croatia
f
4,5
NMR Center, Ruđer Bošković Institute, Bijenička cesta 54, 10000 Zagreb, Croatia
transmitter with a substituted β-phenylethylamine structure.
1
13
†
Electronic supplementary information (ESI) available: File 1: H and C NMR
Therefore, a vast number of β2-agonist drugs possess the
α-aryl-β-amino alcohol fragment in their structure (Fig. 1).
Given the importance of β-amino alcohol moiety in organic
spectra of products, detailed NMR information obtained from 2D NMR spectra
and NOESY spectra of compounds 6 and 10, Cartesian coordinates and com-
puted total energies of complexes, IRC curves of transitions states. File 2: The
normal mode of the transition state of the para-quinone methide – methylamine synthesis, numerous synthetic routes have been developed in
reaction complex leading to β-aryl-β-amino alcohol isomer with imaginary fre-
quency of 139.4i cm . File 3: The normal mode of the transition state of the
order to control the regioselectivity and stereoselectivity of the
−
1
1,2,6
final product.
Among functional group manipulation
bromo alcohol 3 – methylamine reaction complex leading to product B with ima-
ginary frequency of 390.2i cm−1. File 4: The normal mode of the transition state
methods, the formation of amino alcohols via the reaction of
an epoxide and an amine is commonly employed due to the
availability of starting materials. However, the regioselectivity
of the bromo alcohol 3 – methylamine reaction complex leading to product A
with imaginary frequency of 413.2i cm− . See DOI: 10.1039/d0ob02095h
1
7
This journal is © The Royal Society of Chemistry 2020
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 1 Structures of bronchodilating β2-agonists.
depends substantially on the structures of the epoxide and the
amine, as well as on reaction conditions (solvent, presence of
8
acid, base or other catalyst such as metal salts). Nevertheless,
it has been widely used in the synthesis of numerous
9
–15
β-adrenergic agonists,
usually without discussing the regio-
selectivity of the ring opening.
Although the reaction of a β-halohydrin and an amine is
less common, this synthetic approach has also been used for
the preparation of several structurally analogous compounds,
Fig. 2 Structure of trantinterol and new β2-agonists with β-aryl-
β-amino alcohol scaffold.
1
6–19
both using halohyrin with protected,
and unprotected
2
0–23
hydroxyl group.
In the mentioned papers, β-halohydrin
derivative salmeterol, a well-known long-acting β2-adenorecep-
substitution was considered clear and simple, with α-aryl-
β-amino alcohol being the only reported product. However, it
seems that this reaction is not straightforward as expected.
There have been reports on the β-aryl-β-amino alcohol being
tor agonist used as an inhaled bronchodilator for the prevention
31
of bronchospasm (Fig. 1), was pointed out as the β2-agonist
30
with the highest inhibition potency for both cholinesterases.
Here, we demonstrate that both salmeterol (1) and its β-aryl-
β-amino alcohol isomer can be successfully obtained from the
same starting material, depending on the synthetic approach
used. Furthermore, the preparation of α-aryl-β-aminoethanol
moiety is not always straightforward, and structural isomer can
be obtained in the reaction of amine and β-halohydrin.
2
4
formed in the reaction instead of the desired product. In
fact, the synthesis of brombuterol was published using this
2
5
approach, and subsequently the structure of prepared com-
2
6
pound was corrected to its β-aryl-β-amino alcohol isomer.
This nonclassical β-aryl-β-amino alcohol scaffold has been
given the special attention in recent years due to the promising
new bronchodilator isomer of mabuterol – trantinterol, the β2-
adrenoceptor agonist currently being evaluated in clinical
2
. Results and discussion
5
,27
trials.
Based on the success of trantinterol, new potential
bronchodilators with nonclassic β-aryl-β-amino alcohol Retrosynthetic analysis for the preparation of salmeterol start-
scaffold have been developed, some of which have proven to ing from commercially available methyl 5-acetylsalicylate is
2
8,29
be effective β2-adrenergic agonists (Fig. 2).
We previously reported a study on the influence of some β2-
shown in Scheme 1.
In the first synthetic route, methyl 5-acetylsalicylate was
32
agonists on the activity of human acetylcholinesterase (AChE; brominated at the α-position yielding 2, which was reduced
3
0
EC 3.1.1.7) and butyrylcholinesterase (BChE; E.C. 3.1.1.8). We to methyl 5-(2-bromo-1-hydroxy-ethyl)-salicylate (3) in the next
demonstrated that some β2-agonists, derivatives of resorcinol, step (Scheme 2). The reaction of 3 and amine 4 in acetonitrile
catechol or saligenin, reversibly inhibited human acetylcholin- with the help of triethylamine as the base produced product 6.
esterase and butyrylcholinesterase and thus reduced their After reduction, white solid 7 was obtained, with the same
1
activity, affecting their roles in the organism. The saligenin molecular peak in MS, and the same number of signals in H
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Organic & Biomolecular Chemistry
Paper
Scheme 1 Strategy for the preparation of salmeterol.
Scheme 2 Preparation of compound 7.
13
and C NMR spectra as salmeterol. However, a few of the route was analogous to the one previously described in the lit-
3
4
signals in the NMR spectra were shifted when compared with erature, except compound 2 was used instead of 5-(bromoa-
the literature data for salmeterol (ESI†). Therefore, in order to cetyl)-2-hydroxybenzaldehyde (Scheme 3). Amine 5 reacted
resolve this discrepancy, the second synthetic route to prepare readily with 2 to give α-aminoketone 8, which was further
salmeterol was applied.
reduced to produce benzyl protected β-amino alcohol 9. After
Since α-haloketones react with primary amines in an unspe- removal of the protecting group and reduction of 10, com-
3
3
cific way, which results in a complicated mixture of products,
pound 1 was obtained, which was confirmed to be salmeterol
9
,34,35
secondary amine 5 was used in the reaction. This synthetic by comparing its NMR spectra with literature data.
Scheme 3 Preparation of salmeterol (1).
This journal is © The Royal Society of Chemistry 2020
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
1
13
The comparison of H and C NMR spectra of the obtained NMR spectra were used for differentiating isomers. In particu-
compounds 7 and 1 revealed that salmeterol and its isomer lar, the shifts of three hydrogen atoms between the hydroxyl
1
were synthesized, as indicated by MS analysis. Therefore, we and amino group in the H NMR spectrum distinguished con-
analyzed compounds 6 and 10 with more NMR techniques in siderably for these two structural isomers. The difference is
1
3
order to determine the structure of compound 6. These com- also notable for two corresponding signals in the C NMR
pounds were selected for NMR analysis due to their higher spectrum (ESI†). Primary amines BuNH and t-BuNH reacted
2
2
solubility in the range of organic solvents, compared to the same as the long-chain amine used before and produced
reduced compounds 7 and 1. Structures of two compounds predominately compounds 11 and 12, respectively. The lower
were undoubtedly confirmed by 1D and 2D NMR spectra. regioselectivity of BuNH
2
is presumably the consequence of
Different proton and carbon chemical shifts of atoms near the decreased steric effects of this nucleophile. Although BuNH
2
1
amine group are clearly seen from the H and APT NMR shows lower regioselectivity, it was used in further testing due
spectra (see ESI†) and approved by COSY, NOESY, HMQC and to its availability and the fact that the steric effects of the
HMBC spectra. NOESY experiments have been shown as very nucleophile were minimized. Also, the result itself leaves room
useful in gaining more information about compounds 6 and for improvement under other reaction conditions.
1
0. The common NOE intramolecular interactions in both
Since all primary amines reacted in the same way in the
compounds are between the OCH
3
group and OH group on reaction, it was clear that the structure of aryl β-halohydrin
the phenyl ring and between the H-7 atom and phenyl atoms played a dominant role in the direction of the reaction. As
H-3 and H-5 (Fig. 3, ESI†). The two H-8 atoms in compound 10 expected, when the reaction was performed with halohydrins
are almost perpendicular to the phenyl ring plane thus without methyl salicylate moiety, the yield of the reaction was
showing NOE interactions with all phenyl proton atoms (H-3, lower and the isomer ratio approached 1 : 1 (Scheme 5). Also,
H-5 and H-6), while in 6 the interaction with H-6 is missing. in the reactions with 2-bromo-1-phenylethanol (15) and methyl
Long range interaction can be seen in 6 between H-8 and the 3-(2-bromo-1-hydroxyethyl)benzoate (16), the corresponding
OH group on atom C-1. Specific molecule conformation in the epoxide was isolated in moderate yield (39% and 44% in reac-
solvent allows these interactions through space.
tions with 15 and 16, respectively). This suggests that the poss-
To elucidate the reasons for the selectivity obtained in the ible reaction paths are via epoxides, which were formed and
reaction of β-halohydrin and amine, other primary amines isolated under reaction conditions. When the reaction was per-
(
2 2
n-butylamine, BuNH and tert-butylamine, t-BuNH ) were first formed with halohydrin 3, the corresponding epoxide was not
tested (Scheme 4). It was vital to see if the long-chain amine detected. It is also interesting to point out that the epoxide
itself has any influence on the direction of the reaction since generated from benzyl protected compound 3 was previously
3
6
the synthesis of the amine is resource- and time-consuming. used for the preparation of salmeterol. The authors did not
mention that another isomer was generated in the reaction
(
performed in THF). They also attempted the reaction with
bromo alcohol as an electrophile, but a poor yield of the
product was obtained.
Alkyl substituted epoxides are less dependent on reaction
conditions and under basic conditions generate almost exclu-
sively
the
α-alkyl-β-aminoethanol
isomer
via
the
SN2 mechanism. However, aromatic epoxides are much more
sensitive to reaction conditions due to the proximity of the aro-
matic ring which stabilizes the positive charge on the carbon
atom next to it. Therefore, it seemed reasonable to test other
reaction conditions for the halohydrin-amine reaction. To see
if the reaction would proceed in the same way under different
conditions, other solvents and bases were used in the reaction
Fig. 3 NOE interaction (top) and molecular structures (bottom) of
compounds 6 and 10.
2
between bromo alcohol 3 and BuNH (Table 1). The selected
Scheme 4 The reaction of 3 with primary amines.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Organic & Biomolecular Chemistry
Paper
2
Scheme 5 The reaction of aryl β-halohydrins and BuNH .
a
2
Table 1 The reaction of bromo alcohol 3 and BuNH under different conditions
b
c
Entry
Solvent
Base
Time /h
Yield/%
Ratio 11 : 13
1
2
3
4
5
6
7
Acetonitrile
THF
Toluene
Chloroform
Acetone
i-PrOH
Et
Et
Et
Et
Et
Et
Et
3
N
3
N
3
N
3
N
3
N
3
N
3
N
48
72
72
72
48
48
2
83
45
38
82
n.d.
90
96 : 4
41 : 59
63 : 37
82 : 18
41 : 59
100 : 0 e
(100 : 0)
d
e
BuNH
2
(43)
8
9
1
1
1
Acetonitrile
Acetonitrile
Acetonitrile
Acetonitrile
Acetonitrile
—
BuNH
Diethylaniline
DBU
72
24
72
2
57
93
56
<39
(61)
85 : 15
82 : 18
86 : 14
100 : 0
2
0
1
2
f
g
g
K
2
CO
3
2
(100 : 0)
a
b
2
BuNH (1.1 equiv.) was added to a solution of 3 (1 equiv.) in the corresponding solvent (10 mL per 1 mmol of 3) followed by base (2 equiv.).
c
The reactions were processed when HPLC analysis indicated that 3 was no longer present in the reaction mixture or after 72 h. Determined by
d
e
NMR analysis of the obtained product. Not determined due to an unidentified product present in the mixture in high portion. The product is
an amide formed from the corresponding ester. An unidentified product is present in the mixture. The product is the phenolic salt of 11.
f
g
solvents included nonpolar (toluene, chloroform), moderately formed for 72 hours. It must be stressed that the corres-
polar aprotic (THF), polar aprotic (acetone, acetonitrile) and ponding epoxide was not isolated in any of these reactions.
polar protic solvents (i-PrOH, BuNH ). To test the influence of Moreover, the course of the reaction in acetone was very
2
the strength of the base in the reaction, diethylaniline was unclear since one more product, with similar polarity on silica
chosen as a weaker base then Et
chosen as stronger organic and inorganic bases, respectively. of unknown peaks in the NMR spectrum of the product. When
Also, BuNH was used as an example of primary instead of a BuNH was used as the solvent, the product was an amide that
3 2 3
N, and DBU and K CO were gel, is formed in the reaction, as indicated by the appearance
2
2
tertiary amine, and the reaction was performed without the resulted from nucleophilic substitution of amine at the ester
excess of the base. Since the reaction itself is rather slow, the group on the aromatic ring. The best regioselectivity and yield
temperature was kept in the range of 55–60 °C in all sub- was obtained in i-PrOH. Nevertheless, acetonitrile was used as
sequent testings.
a solvent when testing different organic bases in order to elim-
It is evident from the results (Table 1) that the solvent plays inate the possibility of a nucleophilic attack of i-PrOH on a
an important role in the direction of the reaction. Although it substrate when stronger bases were tested.
is clear that polar protic solvents favor isomer 11, the influence
The effect of different bases on the course of the reaction is
of the polarity of the remaining solvents on the direction of more straightforward. Although stronger bases accelerate the
the reaction is ambiguous. The reaction rate was rather low in reaction and direct it toward isomer 11, the yield of the reac-
toluene and chloroform, indicated by reactant 3 retrieved after tion is reduced. This outcome is probably due to side reac-
chromatographic workup even after the reaction was per- tions, which are frequent when strong bases are employed in
This journal is © The Royal Society of Chemistry 2020
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
the combination with elevated temperature. Also, when an in- product when bromo alcohols without an unprotected
organic base was used, the obtained product was the phenolic hydroxyl group on the phenyl ring were used. Furthermore,
salt of the product. When an excess base was not added or a bromo alcohol 3 seems to be quite stable under milder basic
weaker base was used, the 11 : 13 ratio was lower, as well as the conditions since it was isolated in cases when the reaction was
reaction yield, with reactant 3 retrieved after the reaction. not completed after 72 h. Finally, the combination of appropri-
Again, corresponding epoxide was not isolated after the reac- ate leaving group in the proximity of phenol, and the basic
tion. Although the reaction yield with a primary amine as the reaction conditions were reported to trigger the formation of
base was high, the 11 : 13 ratio was decreased compared to para-quinone methides, both as stable products which can be
3
9–41
42
other bases. Thus, it is clear that the choice of base requires isolated,
or as useful reactive intermediates.
fine tuning – it should be rather strong, but not strong enough
to promote side reactions.
To support our hypothesis, we decided to turn to compu-
tational methods. First, we analyzed the nucleophilic attack of
Considering the obtained experimental results, we propose amine (methylamine) on our proposed para-quinone methide
that the reaction between aromatic β-halohydrins and amines intermediate in acetonitrile and aimed to estimate the energy
in basic conditions can take place through three reaction path- barrier (Fig. 4). As initial reactants in this reaction, complex of
ways (Scheme 6). The first two are well-established pathways – para-quinone methide and methylamine (p-QM:MeNH
2
) was
direct S 2 substitution (pathway I) and reaction via epoxide optimized, harmonic analysis was performed, and the absence
N
intermediate (pathway II). The third pathway via para-quinone of imaginary frequencies was confirmed. The nitrogen in this
β
methide (pathway III) we introduce for the first time. minimum energy structure is 4.46 Å away from electrophilic C
Depending on the structure of the β-halohydrin moiety and carbon atom in the vicinity of phenyl ring, and the ester group
the reaction conditions, one of these reaction pathways is pre- on the opposite part of the ring encloses the angle of −44.65°
dominant. In the vast majority of cases the reaction proceeds with the phenyl ring plane. The transition state has one
−1
through the first two pathways. However, when a phenol is normal mode with imaginary frequency (ν = 139.4i cm ). The
present in the structure of aromatic halohydrin, the formation displacement vectors clearly indicate that this is the mode
III
β
of para-quinone methide intermediate (pathway III) is plaus- leading to the intermediary product IM A (ESI†). C carbon
2
ible. This reactive intermediate is then prone to nucleophilic atom is sp hybridized, with small deviation of expected pla-
addition under the above mentioned reaction conditions narity (dihedral angle C3–C4–C7–H is −7.2°). From the point
3
7,38
(Schemes 2 and 4).
2
of view of the initial complex p-QM:MeNH , transition state
III
−1
Besides the formation of β-aryl-β-amino alcohol isomer, TS A is only 5.9 kcal mol higher in energy, while the inter-
there are other experimental indications that suggest the reac- mediary product IM A is more stable for 13.7 kcal mol
III
−1
.
tion does not proceed via conventional pathways. The phenol Final step leading to the reaction product A is the deprotona-
is efficiently deprotonated under reaction conditions, which is tion of the amine and base-assisted hydrogen transfer to the
supported experimentally – in the case when K CO was used phenyl oxygen. In the product, carboxylic group is again in the
2
3
as a base, the potassium salt of 11 was isolated as the product same plane with the phenyl ring, forming hydrogen bond with
Table 1). Also, the corresponding epoxide was not isolated in hydroxyl group from neighboring carbon atom. When calcu-
(
the case of bromo alcohol 3, while it was a significant side lations were performed in non-polar solvent (n-hexane), the
Scheme 6 Three possible reaction pathways between the aromatic bromo alcohol and an amine under basic conditions.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Organic & Biomolecular Chemistry
Paper
Fig. 4 The energy profile for half reaction from para-quinone methide – methylamine reaction complex to β-aryl-β-amino alcohol isomer. The
relevant interatomic distances are in Angstrom, and the dihedral angle C3–C2–C–O is in degrees. B3LYP/aug-cc-pVDZ with SMD solvation model
was used for geometry optimization, followed by B2-PLYP/aug-cc-pVTZ with SMD solvation model (solvent was acetonitrile) single point
calculations.
estimated energy barrier for the attack of methyl amine on the imaginary frequency ν = 390.23i cm− (ESI†). This transition
1
β
−1
−1
2
C carbon atom of p-QM:MeNH was 12.2 kcal mol (ESI†), state is more than 73.3 kcal mol higher in energy then the
which is more than twice the value obtained in acetonitrile. initial complex. The search for lower lying transition states for
This indicates that the reaction in n-hexane is slower by four this mechanism failed. So, supposing that 3 is the dominant
orders of magnitude, which is in agreement with our experi- species in the solution and the reaction proceeds via
mental findings.
Furthermore, we computed the reaction barrier for alterna- to the high barrier, A would not be isolated at all.
tive classical mechanisms – pathways I and II in Scheme 6. The second classical pathway is nucleophilic attack of the
SN2 mechanism, the dominant product would be B, while due
The S 2 reaction mechanism (Fig. 5), where direct attack of amine on the epoxide intermediate (epoxy-3) formed under
N
α
methylamine on the terminal carbon atom C occurs, leads basic conditions (Fig. 6, ESI†). According to our calculations,
toward expected product B. In the minimum energy complex the products of this half reaction would be both isomers A and
of 3 and methylamine, the amine is 3.83 Å away from the term- B, with expected B isomer in surplus. The expected isomer B is
−1
N
inal carbon. The transition state for the S 2 mechanism is thermodynamically more stable than A by 4.1 kcal mol , and
7.3 kcal mol− higher in energy. It is characterized by sp
1
2
the corresponding estimated barrier is lower by 1.6 kcal mol .
−1
1
hybridized terminal carbon, while its neighborhood can be If this mechanism was dominant, the product ratio would be in
approximated by trigonal bipyramid, with leaving bromine disagreement with experimental observations. Furthermore, the
and attacking methylamine on axial positions. Calculations for analysis of analogous reaction pathway with compound 16 and
investigation of analogous mechanism, but with amine attack- methylamine shows similar energy profile (ESI†) which is in
ing the chiral carbon atom, resulted in interesting transition agreement with experimental results obtained for 16
β
state. The bond between amine’s nitrogen and C carbon is (Scheme 5). This further corroborates the assumption that the
formed. Carbon atom remains in the center of the tetrahedron. presence of hydroxyl group on the aromatic ring in 3 directs the
Terminal carbon Cα to bromine distance of 2.52 Å indicates reaction towards different intermediate than in the case of 16.
that bromine is leaving the molecule. But the most interesting
To summarize, the outcome of the reaction between the
β
part is transfer of the hydroxyl moiety from C toward terminal aryl β-halohydrin 3 and an amine is both solvent and base
carbon atom, leading to the product A. The oxygen is closer to dependent, with the β-aryl-β-amino alcohol isomer being iso-
α
β
C (2.28 Å) than to the C (2.54 Å) carbon. This is corroborated lated in most cases. The outcome of the reaction cannot be
by analysis of the displacement vectors of normal mode with explained by two conventional mechanisms – direct S 2 or the
N
This journal is © The Royal Society of Chemistry 2020
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 5 Initial (left) and transition state (right) geometries relevant for S
N
2 mechanism. Methylamine attacking terminal Cα (top) and Cβ (bottom)
carbon atom, with relevant distances in Angstrom.
Fig. 6 The energy profile for half reaction from epoxide – methylamine reaction complex. The relevant interatomic distances are in Angstrom.
B3LYP/aug-cc-pVDZ with SMD solvation model was used for geometry optimization, followed by B2-PLYP/aug-cc-pVTZ with SMD solvation model
(
solvent was acetonitrile) single point calculations.
epoxide pathway. We have shown that the barrier for concerted temperature of 330 K. Thus, the expected product B is both
α
methylamine attack on C and leaving of bromine anion is thermodynamically and kinetically favored. The presence of
−
1
2
β
around 17 kcal mol , with expected final α-aryl-β-amino para-quinone methide intermediate with sp hybridized C
β
alcohol isomer, while the barrier for attack on C is signifi- atom prone to the attack by amine results exclusively in A-like
cantly higher. On the other hand, the estimated barriers for product, i.e. β-aryl-β-amino alcohol, and fully explains the
α
β
methylamine attack on C and C of the corresponding regioselectivity observed experimentally.
epoxide are very close in energy. The estimated barrier of Finally, we were able to synthesize two isomers of the sali-
.6 kcal mol− leads to a reaction rate difference of 1 : 11.5 at a genin class of β2-agonists bronchodilators. The isomer of sal-
1
1
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Organic & Biomolecular Chemistry
Paper
tate 70 to glycine (D70G), a residue located in the peripheral
site of the active centre of BChE and close to the entrance of
the active centre gorge. The D70G mutant was confirmed to
have a different conformation compared to usual BChE, par-
ticularly affecting the tryptophane 82 that participates in
accepting the ligand and its proper orientation at the entry of
4
4
the active site of the BChE.
Generally speaking, the inhibition profile of the tested
isomers and cholinesterases is similar to that reported for
albuterol and salmeterol; inhibition potency is lower toward
AChE compared to BChE and the highest inhibition potency is
that for usual BChE. When the inhibition potency of isomers
is compared to that of corresponding bronchodilators, isomers
are 2–5 times more potent inhibitors. It seems that structural
rearrangements of the hydroxyl group in the alkyl chain on the
benzene ring, as in isomers of albuterol and salmeterol, can
ensure an orientation in the active site of a cholinesterase that
is more favourable for inhibition compared to that of the
corresponding bronchodilators. The exception is the inhibition
of usual BChE by iso-albuterol, which was twice less potent
compared to that by albuterol suggesting that rearrangement
of the hydroxyl groups present in albuterol is more optimal for
inhibition.
Fig. 7 Isomers of saligenin type β2-agonist bronchodilators.
meterol is compound 7, and the reduction of compound 12
afforded 17, the isomer of albuterol (Fig. 7). These isomers
were tested for inhibition of acetylcholinesterase (AChE; EC
3
.1.1.7) and butyrylcholinesterase (BChE; E.C. 3.1.1.8) since in
3
0
our earlier paper we demonstrated that albuterol and salme-
terol can reduce human cholinesterase activity. Since the
efficacy of compound that targets BChE activity can be affected
1
1,30
by the human BChE gene polymorphism,
inhibition
activity was tested towards two BChE variants with different
catalytic properties, usual BChE (hydrolyzes certain short-
acting neuromuscular blocking agents widely used during
anesthesia, i.e. succinylcholine and mivacurium) and atypical
BChE (cannot hydrolyze succynilcholine).
3. Conclusion
Kinetic studies revealed that iso-albuterol and iso-salme-
terol reversibly inhibited all of the tested cholineaterases with
enzyme-inhibitor dissociation constants (K ) ranging between
We developed two synthetic routes that enable preparation of
both saligenin-based β2-agonists, albuterol and salmeterol,
and their isomers, iso-albuterol and iso-salmeterol. All pro-
ducts are obtained in high yield starting from methyl 5-acetyl-
salicylate. HRMS analysis confirmed that the prepared com-
pounds are isomers, while 1D and 2D NMR were used for
determining the structure of the prepared isomers. The
obtained structural isomers contain β-aryl-β-aminoethanol
moiety opposed to the substituted α-aryl-β-aminoethanol struc-
ture found in common β2-adenoreceptor agonists. The study
of the halohydrin-amine reaction, responsible for the for-
mation of the isomer, revealed that the methyl salicylate
moiety in the structure of β-halohydrin dictates the reaction
path, while the effect of the used primary amine is negligible.
Also, the solvent plays a relevant role in the direction of the
reaction with polar protic solvents favoring the unexpected
isomer in the reaction. On the other hand, the base has more
straightforward effect on the rate of the reaction and on the
products’ isomer ratio. Fine tuning of the base strength is
essential for adequate reaction yield and isomer ratio, since
strong bases lower the yield, while weaker bases increase reac-
tion time and shift isomer ratio toward expected isomer. The
most probable reaction path in the presence of the strong
base, via para-quinone methide intermediate, is corroborated
by quantum chemistry calculations. This is in accordance with
experimental findings, and explains the influence of base
strength, both on reaction mechanism and reaction outcome.
Since we successfully obtained isomers of albuterol and salme-
terol, the inhibition potency of iso-albuterol and iso-salmeterol
(
I)
2
.6 µM and 1.3 mM (Table 2) classifying these compounds as
4
3
low-potency inhibitors. The highest inhibition potency was
displayed by iso-salmetrol toward usual BChE, while the lowest
was that of iso-albuterol toward AChE. The inhibition potency
of iso-salmetrol was 40–80 times higher than that of iso-albu-
terol, which could have been due to the elongation of the tert-
butylamino group, as it is in iso-albuterol, to phenylbutoxy-
hexylamino group present in iso-salmeterol. Both compounds
are seven times more potent inhibitors of BChE than of AChE,
meaning that both compounds can be considered as selective
BChE inhibitors. The approximately five-fold decrease of inhi-
bition potency toward atypical BChE compared to usual BChE
could have been a consequence of the point mutation of aspar-
Table 2 Reversible inhibition of cholinesterases by bronchodilating β2-
agonists bearing the saligenin structure
a
K
(I) /mM
AChE
Usual BChE
Atypical BChE
Iso-albuterol (17)
Iso-salmeterol (7)
1.3 ± 0.2
0.016 ± 0.003
2.0 ± 0.2
0.17 ± 0.02
0.69 ± 0.13
0.017 ± 0.002
1.4 ± 0.2
0.0026 ± 0.0003
0.071 ± 0.006
0.013 ± 0.001
3
0
Albuterol
3
0
Salmeterol
0.030 ± 0.002
0.086 ± 0.018
a
The enzyme-inhibitor complex dissociation constant (K(I) ± SE) was
determined by linear regression from Ki,app constants obtained from at
least two experiments.
This journal is © The Royal Society of Chemistry 2020
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
f
toward human acetylcholinesterase (AChE) and usual and aty- chromatography (DCM, R = 0.25) to give a colourless oil which
pical butyrylcholinesterase (BChE) was tested and compared to crystallizes (0.64 g, 91%).
1
the inhibition potency of common saligenin-based β2-ago-
3
H NMR (300 MHz, CDCl ) δ 10.77 (1H, s), 7.87 (1H, d, J =
nists. Kinetic studies demonstrated that both isomers are low- 2.3 Hz), 7.47 (1H, dd, J = 8.5, 2.3 Hz), 6.99 (1H, d, J = 8.5 Hz),
potency reversible inhibitors of human cholinesterases with 4.87 (1H, dt, J = 8.7, 2.8 Hz), 3.95 (3H, s), 3.60 (1H, dd, J = 10.4,
an approximately seven times higher preference toward BChE 3.6 Hz), 3.51 (1H, dd, J = 10.4, 8.7 Hz), 2.69 (1H, d, J = 2.8 Hz);
1
3
compared to AChE.
C NMR (75 MHz, CDCl
3
) δ 170.30, 161.66, 133.41, 131.19,
1
27.61, 118.10, 112.39, 73.08, 52.49, 40.09. HRMS (ESI-TOF) m/
+
z: [M + Na] calcd for C13
-(4-Phenyl-butoxy)-hexylamine (4). Potassium phthalimide
0.65 g, 3.5 mmol) was added to a solution of [4-(6-bromo-hexy-
loxy)-butyl]-benzene (1.00 g, 3.2 mmol) in DMF (10 mL). The
.1.1. General considerations. All of the reactions were con- reaction mixture was stirred at 60 °C for 2 hours. After cooling
3
H20NO Na 261.1341; found 261.1350.
6
4
. Experimental section
(
4
.1. Chemistry
4
ducted under an argon atmosphere unless otherwise noted. to room temperature water (10 mL) was added and extracted
THF was distilled from lithium aluminum hydride. All of the with diethyl ether (3 times). The combined organic extracts
other reagents and solvents were purchased from commercial were extracted with saturated aqueous NaHCO
3
solution,
sources and used without purification. TLC was performed on washed with water, dried over Na SO and concentrated under
2
4
aluminium backed silica plates (60 F254, Merck). UV light reduced pressure to afford a colourless oil. The oil was dis-
254 nm) or phosphomolibdic acid reagent were used for solved in EtOH (15 mL) and hydrazine hydrate (50–60%,
visualizing. Column chromatography was performed on silica 0.3 mL) was added. The reaction mixture was refluxed for
(
1
13
gel (Silica Gel 60, 70–230 mesh, Fluka). H and C NMR were 1.5 hours. The solvent was evaporated and diethyl ether was
recorded on a Bruker AV 300 spectrometer, while COSY, added to the white residue. The mixture was filtered and white
NOESY, HMQC and HMBC spectra were recorded on a Bruker solid washed with diethyl ether. The filtrate was concentrated
4
5
AV 600 spectrometer. The NOESY spectra were measured in a under reduced pressure to give colourless oil (0.75 g, 92%).
phase-sensitive mode, with a mixing time of 0.90 s and 32
1
3
H NMR (300 MHz, CDCl ) δ 7.30–7.23 (2H, m), 7.20–7.15
scans per each increment. The spectral width was 9615.38 Hz, (3H, m), 3.43–3.35 (4H, m), 2.74–2.67 (2H, m), 2.66–2.59 (2H,
1
3
2
048 points in the F2 dimension and 512 increments in the F1 m), 2.55 (2H, br s), 1.72–1.42 (8H, m), 1.40–1.30 (4H, m);
C
3
dimension, subsequently zero-filled to 1024 points. The result- NMR (75 MHz, CDCl ) δ 142.51, 128.42, 128.26, 125.67, 70.85,
ing FID resolution was 4.69 Hz per point and 18.75 Hz per 70.71, 41.84, 35.74, 32.96, 29.71, 29.41, 28.09, 26.72, 26.04;
+
point in the F2 and F1 dimensions, respectively. Chemical MS/+ESI (m/z): 150.3 ([C16
H27NO + H] ).
shifts (δH and δC) are quoted in parts per million (ppm), refer- 4.1.3. Synthesis of salmeterol (1). The synthesis was analo-
3
4
enced to TMS. HPLC measurements were done using a gous to that described in literature, apart from using methyl
Shimadzu 10A VP HPLC system. Nucleosil 100-5 C18, 250 mm 5-(bromoacetyl)-salicylate as an electrophile.
×
4.6 mm column, A: H
2
O, MeOH, H
3
PO
4
(85%) = 90 : 10 : 0.5,
Methyl
5-(2-{benzyl-[6-(4-phenyl-butoxy)-hexyl]-amino}-1-
B: MeOH, Gradient method: 10/100/100/10%B in 0/20/25/ hydroxy-ethyl)-salicylate (9). To a solution of 2 (0.61 g,
7 min, 1 mL min− , 220 nm. The LC/MS system consisted of 2.2 mmol) in acetonitrile (10 mL), the solution of 5 (0.76 g,
1
2
an Agilent Technologies 1200 Series HPLC instrument and 2.2 mmol) in acetonitrile (15 mL) was added dropwise under
Agilent Technologies 6420 Triple Quad LC/MS equipped with inert atmosphere. Triethylamine (0.62 mL, 4.4 mmol) was
electrospray ion source. High resolution mass spectrometry added dropwise and the reaction mixture was stirred at room
(
HRMS) was performed on a 4800 Plus MALDI TOF/TOF temperature for 40 min. The solvent was evaporated, DCM
3
4
Analyzer. [4-(6-Bromo-hexyloxy)-butyl]-benzene,
methyl 5- (25 mL) added to the resulting yellow residue and washed with
SO and concen-
were prepared according to known trated under reduced pressure to give a yellow oil. Further puri-
3
2
(bromoacetyl)-salicylate (2) and N-benzyl-6-(4-phenylbutoxy)- water. The organic layer was dried over Na
2
4
3
4
1
-hexanamine (5)
procedures.
.1.2. Synthesis of amine and β-halohydrin
Methyl 5-(2-bromo-1-hydroxy-ethyl)-salicylate (3). Methyl 5-
fication of gained product 8 was unsuccessful, therefore the
second reaction step was performed with crude product.
The solution of obtained oil (1.38 g) in MeOH (35 mL) was
4
(bromoacetyl)salicylate (0.70 g, 2.6 mmol) was suspended in cooled in ice bath and sodium borohydride (100 mg,
MeOH (25 mL) and the mixture was cooled to 0 °C. Sodium 2.6 mmol) was added in several portions. After addition, the
borohydride (99 mg, 2.6 mmol) was added in several portions. reaction mixture was stirred at room temperature for 1 hour.
After addition the reaction mixture was stirred at room temp- The solvent was evaporated and water (15 mL) was added to
erature for 1 hour. The solvent was evaporated and saturated the resulting white solid and extracted with DCM (3 times).
aqueous NH
white solid and extracted with DCM (3 times). The combined concentrated under reduced pressure to afford a yellow
organic extracts were washed with water, dried over Na SO residue. The crude product was purified by silica gel column
4 2 4
Cl solution (15 mL) was added to the resulting The combined organic extracts were dried over Na SO and
2
4
f
and concentrated under reduced pressure to afford a colour- chromatography (DCM–MeOH, 50 : 1 to 20 : 1, R = 0.64 for
less oil. The crude product was purified by silica gel column 20 : 1) to give a colourless oil (0.77 g, 65%).
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Organic & Biomolecular Chemistry
Paper
1
H NMR (300 MHz, CDCl
3
) δ 10.69 (1H, s), 7.78 (1H, d, J =
Methyl 5-{2-hydroxy-1-[6-(4-phenyl-butoxy)-hexylamino]-ethyl}-
2
8
.1 Hz), 7.38–7.23 (8H, m), 7.19–7.15 (3H, m), 6.94 (1H, d, J = salicylate (6). Yellow oil (0.35 g, 80%) starting from 3 (0.28 g).
.6 Hz), 4.58 (1H, dd, J = 10.0, 3.7 Hz), 3.93 (3H, s), 3.88 (1H, d,
R
f
= 0.42 (DCM–MeOH, 9 : 1).
H NMR (300 MHz, CDCl ) δ 7.99 (1H, d, J = 2.3 Hz), 7.77
3
1
J = 13.4 Hz), 3.49 (1H, d, J = 13.4 Hz), 3.39 (4H, m), 2.66–2.40
13
(
(
1
1
2
6H, m), 1.71–1.48 (9H, m), 1.34–1.25 (4H, m); C NMR (1H, dd, J = 8.6, 2.3 Hz), 7.32–7.25 (2H, m), 7.21–7.16 (3H, m),
75 MHz, CDCl ) δ 170.58, 161.05, 142.58, 138.62, 133.53, 7.07 (1H, d, J = 8.6 Hz), 4.32–4.25 (1H, m), 4.14–4.05 (1H, m),
33.14, 129.08, 128.52, 128.49, 128.33, 127.35, 125.75, 117.67, 3.99 (3H, s), 3.98–3.90 (1H, m), 3.39 (2H, t, J = 6.3 Hz), 3.35
12.10, 70.91, 70.80, 68.82, 62.55, 58.65, 53.89, 52.31, 35.82, (2H, t, J = 6.5 Hz), 2.75–2.67 (2H, m), 2.63 (2H, t, J = 7.3 Hz),
3
1
3
9.80, 29.49, 28.16, 27.23, 26.95, 26.17.
1.83–1.73 (2H, m), 1.70–1.49 (6H, m), 1.36–1.26 (4H, m);
C
3
Methyl 5-{2-[6-(4-phenyl-butoxy)-hexylamino]-1-hydroxy-ethyl}- NMR (75 MHz, CDCl ) δ 170.02, 162.08, 142.44, 135.18, 130.16,
salicylate (10). Bn protected aminoalcohol
9
(0.23 g, 128.40, 128.26, 125.68, 125.12, 118.86, 112.75, 70.73, 70.64,
0
.43 mmol) was dissolved in MeOH (8 mL) and hydrogenated 64.28, 64.09, 52.56, 46.35, 35.71, 29.55, 29.34, 28.01, 27.28,
+
in the presence of 10% Pd–C (30 mg) for 2 hours. The catalyst 26.78, 25.76; MS/+ESI (m/z): 444.4 ([C26
was filtered over a Celite pad and the filtrate was evaporated Methyl 5-(1-butylamino-2-hydroxy-ethyl)-salicylate (11). Yellow
under reduced pressure to afford a colourless oil which crystal- oil (83 mg, 83%) starting from 3 (103 mg). R = 0.40 (DCM–
lizes (0.19 g, 98%). MeOH, 9 : 1). The product containes 3% of isomer 13 (deter-
H NMR (300 MHz, CDCl ) δ 7.84 (1H, d, J = 2.1 Hz), 7.45 mined by NMR).
5
H37NO + H] ).
f
1
3
1
(
1H, dd, J = 8.6, 2.1 Hz), 7.28–7.25 (2H, m), 7.18–7.15 (3H, m),
.96 (1H, d, J = 8.6 Hz), 4.68 (1H, dd, J = 9.3, 3.3 Hz), 3.94 (3H, (1H, dd, J = 8.6, 2.2 Hz), 7.05 (1H, d, J = 8.6 Hz), 4.32 (1H, m),
s), 3.41 (2H, t, J = 6.4 Hz), 3.38 (2H, t, J = 6.7 Hz), 2.86 (1H, dd, 4.13 (1H, m), 3.97 (3H, s), 3.96–3.91 (1H, m), 2.71 (2H, m), 1.76
3
H NMR (300 MHz, CDCl ) δ 7.99 (1H, d, J = 2.2 Hz), 7.78
6
1
3
J = 12.2, 3.3 Hz), 2.72–2.65 (2H, m), 2.65–2.61 (3H, m), (2H, m), 1.37–1.25 (2H, m), 0.86 (3H, t, J = 7.4 Hz); C NMR
1
1
1
1
2
.71–1.64 (2H, m), 1.63–1.59 (2H, m), 1.57–1.48 (4H, m), (75 MHz, CDCl
3
) δ 170.06, 162.26, 135.30, 130.37, 124.67,
1
3
.37–1.32 (4H, m); C NMR (75 MHz, CDCl ) δ 170.53, 161.10, 119.01, 112.90, 64.27, 64.20, 52.63, 46.18, 29.17, 20.18, 13.63;
3
+
42.57, 133.46, 133.35, 128.47, 128.32, 127.26, 125.74, 117.72, MS/+ESI (m/z): 268.2 ([C14
12.17, 70.88, 70.80, 70.72, 56.96, 52.32, 49.39, 35.80, 29.83, Methyl 5-(1-tert-butylamino-2-hydroxy-ethyl)-salicylate (12).
9.76, 29.47, 28.14, 27.10, 26.15; MS/+ESI (m/z): 444.4 Yellow oil (166 mg, 78%) starting from 3 (219 mg). R = 0.37
4
H21NO + H] ).
f
+
([C26
H
37NO
5
+ H] ).
(DCM–MeOH, 9 : 1).
1
Salmeterol (1). Lithium aluminum hydride (50 mg) was sus-
3
H NMR (300 MHz, CDCl ) δ 8.04 (1H, d, J = 2.3 Hz), 7.87
pended in dry diethyl ether (3 mL) in inert atmosphere. The (1H, dd, J = 8.7, 2.3 Hz), 7.04 (1H, d, J = 8.7 Hz), 4.39 (1H, dd, J
solution of ester 10 (0.31 g, 0.69 mmol) in dry diethyl ether = 9.1, 3.8 Hz), 4.03 (1H, dd, J = 12.1, 9.1 Hz), 3.97 (3H, s), 3.79
1
3
(
8 mL) was added dropwise. The reaction mixture was stirred (1H, dd, J = 12.1, 3.8 Hz), 1.32 (9H, s); C NMR (75 MHz,
at room temperature for 1 hour. The reaction was quenched CDCl ) δ 169.98, 161.97, 135.35, 129.86, 126.75, 118.80, 112.74,
3
with a few drops of water followed by the addition of EtOH 64.70, 61.06, 58.14, 52.58, 28.29; MS/+ESI (m/z): 268.2
+
(
(
5 mL) and saturated aqueous solution of Rochelle salt ([C14
10 mL). The layers were separated and the aqueous phase
H
21NO
4
+ H] ).
4.1.5. Reaction of n-butylamine with other halohydrins.
washed three times with EtOAc. The combined organic extracts The reactions of 2-bromo-1-phenylethanol (15) and methyl 3-
were dried over Na SO and concentrated under reduced (2-bromo-1-hydroxyethyl)benzoate (16) with n-butylamine were
pressure. The product precipitated in cold EtOAc to give an off- conducted according to the above presented reaction
2
4
white solid (0.25 g, 87%).
conditions.
1
H NMR (300 MHz, CDCl
3
) δ 7.33–7.25 (2H, m), 7.22–7.16
4.1.6. Testing different base/solvent conditions. N-
(3H, m), 7.06 (1H, d, J = 8.1 Hz), 6.92 (1H, s), 6.76 (1H, d, J = 8.1 Butylamine (1.1 equiv.) was added dropwise under inert atmo-
Hz), 4.69 (2H, s), 4.56–4.41 (4H, m), 3.47–3.36 (4H, m), 2.70–2.47 sphere to a solution of 3 (1 equiv.) in solvent according to
13
(
6H, m), 1.74–1.40 (8H, m), 1.37–1.26 (4H, m); C NMR Table 2 (10 mL per 1 mmol of bromide). The corresponding
(
75 MHz, CDCl ) δ 155.80, 142.54, 133.66, 128.49, 128.34, 126.59, base (2 equiv.) was added dropwise and the reaction mixture
3
125.97, 125.76, 125.63, 116.38, 71.49, 70.95, 70.84, 63.45, 56.70, was stirred at 55–60 °C until the completion of reaction (moni-
49.40, 35.80, 29.70, 29.63, 29.42, 28.11, 27.14, 26.13.
tored by HPLC) or for 72 hours. The solvent was evaporated
4
.1.4. Reactions of methyl 5-(2-bromo-1-hydroxy-ethyl)-sali- and the obtained residue purified by silica gel column chrom-
cylate with amines
atography (DCM–MeOH, 20 : 1 to 9 : 1) to afford the product.
1
General procedure. The corresponding amine (1 equiv.) was The product was analysed by H NMR spectroscopy to deter-
added dropwise under inert atmosphere to a solution of 3 (1 mine the 11 : 13 ratio.
equiv.) in acetonitrile (10 mL per 1 mmol of bromide).
N-Butyl-5-(1-butylamino-2-hydroxyethyl)-2-hydroxy-benzamide
Triethylamine (2 equiv.) was added dropwise and the reaction (Table 2, entry 7). Yellow oil (49 mg, 43%) obtained after reac-
mixture was stirred at 55 °C (45 °C in the case of tert-butyla- tion in BuNH , starting from 3 (102 mg). R = 0.35 (DCM–
2
f
mine) for 48 hours. The solvent was evaporated and the MeOH, 9 : 1).
1
obtained residue purified by silica gel column chromatography
DCM–MeOH, 20 : 1 to 9 : 1) to afford the product.
H NMR (300 MHz, CDCl
3
) δ 8.56 (1H, s), 7.73 (1H, t, J = 5.3
(
Hz), 7.35 (1H, dd, J = 8.5, 1.3 Hz), 6.97 (1H, d, J = 8.5 Hz), 4.38
This journal is © The Royal Society of Chemistry 2020
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
(
(
=
1H, dd, J = 9.2, 3.6 Hz), 4.23 (1H, dd, J = 12.2, 9.2 Hz), 4.05 4.2. Computational details
1H, dd, J = 12.2, 3.6 Hz), 3.43 (2H, q, J = 6.4 Hz), 2.82 (2H, t, J
8.0 Hz), 1.81 (2H, m), 1.67 (2H, m), 1.46–1.25 (4H, m), 0.92
46
All structures were optimized using B3LYP hybrid functional
and Dunning’s correlation consistent basis sets augmented
1
3
(3H, t, J = 7.3 Hz), 0.86 (3H, t, J = 7.4 Hz); C NMR (75 MHz,
47
with diffuse functions (aug-cc-pVDZ). Solvation effects were
incorporated using Truhlar’s SMD solvation model, while the
solvent was acetonitrile (ε = 35.688). For each structure the
r
harmonic vibrational analysis was performed, to properly
identify the nature of the optimized structures. Single point
calculations for all relevant structures were performed with
CDCl ) δ 169.53, 162.86, 134.42, 126.58, 120.76, 119.34, 115.68,
48
3
6
1
4.86, 63.57, 46.05, 39.82, 31.42, 28.09, 20.38, 20.05, 13.87,
+
3.56; MS/+ESI (m/z): 309.2 ([C17
Potassium 4-(1-butylamino-2-hydroxyethyl)-2-methoxycarbonyl-
phenolate (Table 2, entry 12). Yellow solid (68 mg, 61%)
obtained after reaction with K CO as the base, starting from 3
28 2 3
H N O + H] ).
2
3
49
double hybrid B2-PLYP functional combined with Grimme’s
(
100 mg). R = 0.20 (DCM–MeOH, 9 : 1).
50,51
f
D3BJ dispersion
and triple zeta valence quality basis set
1
3
H NMR (300 MHz, CDCl ) δ 7.76 (1H, d, J = 2.2 Hz), 7.41
(
aug-cc-pVTZ). Solvent effects were included in geometry
(
3
1H, dd, J = 8.6, 2.2 Hz), 6.96 (1H, d, J = 8.6 Hz), 3.94 (3H, s),
.75–3.64 (2H, m), 3.58–3.50 (1H, m), 2.48 (2H, m), 1.50–1.40
optimization and frequency calculations, as well as in all
single point energy calculations. For all calculations Gaussian
13
(
(
1
2H, m), 1.37–1.23 (2H, m), 0.87 (3H, t, J = 7.3 Hz); C NMR
75 MHz, CDCl ) δ 170.51, 161.18, 134.72, 131.45, 128.66,
18.06, 112.42, 66.48, 63.89, 52.45, 47.08, 32.24, 20.48, 14.04;
52
16
was used.
3
+
4.3. Enzyme inhibition
MS/+ESI (m/z): 268.2 ([C14
H
21NO
4
+ H] ); MS/−ESI (m/z): 265.9
−
([C14
H
20NO
4
] ).
Human native serum was used as a source of BChE, while
4
.1.7. Reduction of prepared methyl aminoalcohol recombinant human AChE was used as a source of AChE.
salicylates
BChE variants were confirmed earlier as described pre-
5
3
General procedure. Lithium aluminum hydride (1 equiv.) was viously. The usual BChE was collected from female donors at
suspended in dry THF (4 mL per 1 mmol of ester) in inert atmo- the Institute for Medical Research and Occupational Health,
sphere. The solution of the corresponding ester (1 equiv.) in dry Zagreb, Croatia (IMROH) and the atypical BChE was a gift
THF (8 mL per 1 mmol of ester) was added dropwise. The reac- from Dr Oksana Lockridge (University of Nebraska Medical
tion mixture was stirred at room temperature for 1 hour. The Centre, Eppley Institute, Omaha, USA). Recombinant human
reaction was quenched with a few drops of water followed by AChE was a gift from Dr Florian Nachon (Département de
the addition of EtOH and saturated aqueous solution of Toxicologie, IRBA, France). Enzyme dilutions were done in
Rochelle salt. After stirring for 1 hour, EtOAc was added. The phosphate buffer and those of recombinant AChE with the
layers were separated and the aqueous phase washed two times addition of 0.01% BSA. This study was reviewed and approved
with EtOAc. The combined organic extracts were dried over by the IMROH Ethics Committee. Iso-albuterol was dissolved
Na SO and concentrated under reduced pressure.
in water and iso-salmeterol in ethanol. All further dilutions
2
4
Iso-salmeterol (7). Off-white viscous oil which crystallizes were done in water. All of the experiments were done in 0.1 M
upon standing (0.219 g, 80%) starting from 6 (0.292 g). phosphate buffer, pH 7.4 at 25 °C, the Tecan Infinite
Purified by column chromatography in DCM–MeOH, 9 : 1 + M200PRO plate reader (Tecan, Austria, GmbH, Salzburg,
1
% Et
3
N (R
f
= 0.21). mp 36–37 °C.
Austria). The inhibition mixture (300 µL final) contained a
) δ 7.28–7.24 (2H, m), 7.18–7.14 buffer, enzyme, DTNB (0.33 mM final) and inhibitor, and after
1
H NMR (300 MHz, CDCl
3
(
3H, m), 7.03 (1H, dd, J = 7.9, 1.7 Hz), 6.99 (1H, s), 6.78 (1H, d, the addition of the substrate, the activity was assayed by
5
4
55
J = 7.9 Hz), 4.72 (2H, s), 4.45 (3H, br s), 3.66–3.59 (2H, m), Ellman’s method as described earlier. Acetylthiocholine
.54–3.49 (1H, m), 3.40 (2H, t, J = 6.6 Hz), 3.35 (2H, t, J = 6.6 (ATCh) was used as a substrate for AChE and propyonylthio-
3
Hz), 2.62 (2H, t, J = 7.5 Hz), 2.52–2.41 (2H, m), 1.70–1.63 (2H, choline (PTCh) as a substrate for BChE. Baseline activity was
m), 1.63–1.57 (2H, m), 1.53–1.42 (4H, m), 1.29–1.23 (4H, m); determined in the presence of water since the proportion of
1
3
C NMR (75 MHz, CDCl
3
) δ 155.79, 142.53, 131.04, 128.48, ethanol in AChE and BChE activity measurement was up to
1
6
2
28.35, 128.28, 126.86, 126.03, 125.77, 116.55, 70.94, 70.83, 0.012% and 0.03%, respectively, which does not affect enzyme
6.06, 63.92, 63.58, 47.19, 35.79, 29.68, 29.58, 29.40, 28.07, activity. The inhibition potency of the compounds was deter-
+
7.08, 26.05. HRMS (MALDI TOF/TOF) m/z: [M + H] calcd for mined as a dissociation constant for an enzyme-inhibitor
C H NO 416.2801; found 416.2810.
complex, K(I), evaluated from the impact of substrate concen-
2
5
37
4
Iso-albuterol (17). Yellow solid (92 mg, 64%) starting from 12 trations [S] on the degree of inhibition applying the Hunter–
5
6
(
160 mg). Purified by recrystallization from diisopropyl ether. Downs equation. Detailed description in ref. 55
mp 51–52 °C.
vi
KðIÞ
1
Ki;app ¼
ꢁ ½Iꢂ ¼ KðIÞ
þ
ꢁ ½Sꢂ
H NMR (300 MHz, DMSO) δ 7.25 (1H, s), 7.03 (1H, d, J =
.2 Hz), 6.65 (1H, d, J = 8.2 Hz), 4.45 (2H, s), 3.70 (1H, m), 3.27
v0 ꢀ vi
K
ðSÞ
8
1
3
(1H, m), 3.07 (1H, m), 0.92 (9H, s); C NMR (75 MHz, DMSO) where Ki,app is an apparent enzyme-reversible inhibitor
δ 152.66, 135.20, 127.66, 126.15, 125.93, 114.03, 67.28, 58.78, complex dissociation constant obtained at a given substrate
+
5
8.46, 50.52, 30.01. HRMS (MALDI TOF/TOF) m/z: [M + H]
concentration [S] in the presence of inhibitor concentration [I]
calcd for C H NO 240.1600; found 240.1604.
and calculated from the enzyme activities v and v measured
1
3
21
3
0
i
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Organic & Biomolecular Chemistry
Paper
without inhibitor and in the presence of inhibitor concen- 13 G. Xing, L. Pan, C. Yi, X. Li, X. Ge, Y. Zhao, Y. Liu, J. Li,
5
5
tration [I], respectively.
A. Woo, B. Lin, Y. Zhang and M. Cheng, Bioorg. Med.
Chem., 2019, 27, 2306–2314.
1
4 M. Stanek, L.-P. Picard, M. F. Schmidt, J. M. Kaindl,
H. Hübner, M. Bouvier, D. Weikert and P. Gmeiner, J. Med.
Chem., 2019, 62, 5111–5131.
Conflicts of interest
There are no conflicts of interest to declare.
15 C. Yi, G. Xing, S. Wang, X. Li, Y. Liu, J. Li, B. Lin, A.
Y.-H. Woo, Y. Zhang, L. Pan and M. Cheng, Bioorg. Med.
Chem., 2020, 28, 115178.
1
6 P. A. Glossop, C. A. L. Lane, D. A. Price, M. E. Bunnage,
R. A. Lewthwaite, K. James, A. D. Brown, M. Yeadon,
C. Perros-Huguet, M. A. Trevethick, N. P. Clarke,
R. Webster, R. M. Jones, J. L. Burrows, N. Feeder,
S. C. J. Taylor and F. J. Spence, J. Med. Chem., 2010, 53,
6640–6652.
Acknowledgements
The authors gratefully acknowledge Vladimir Vinković, PhD for
useful comments and discussion. This work was supported by
the Croatian Science Foundation (IP-2013-11-4307). Authors
acknowledge University of Zagreb, University Computing
Centre (SRCE) for granting computational time on ISABELLA 17 A. Liu, L. Huang, Z. Wang, Z. Luo, F. Mao, W. Shan, J. Xie,
cluster.
K. Lai and X. Li, Bioorg. Med. Chem. Lett., 2013, 23, 1548–
552.
1
1
8 R. M. McKinnell, U. Klein, M. S. Linsell, E. J. Moran,
M. B. Nodwell, J. W. Pfeiffer, G. R. Thomas, C. Yu and
J. R. Jacobsen, Bioorg. Med. Chem. Lett., 2014, 24, 2871–
2876.
References
1
2
S. C. Bergmeier, Tetrahedron, 2000, 56, 2561–2576.
O. K. Karjalainen and A. M. P. Koskinen, Org. Biomol. 19 L. H. Jones, J. Burrows, N. Feeder, P. Glossop, K. James,
Chem., 2012, 10, 4311.
U. V. S. Reddy, M. Chennapuram, C. Seki, E. Kwon,
Y. Okuyama and H. Nakano, Eur. J. Org. Chem., 2016, 4124–
R. M. Jones, A. S. Kenyon, S. Patel, D. F. Roberts,
M. D. Selby, R. S. Strang, E. F. Stuart, M. A. Trevethick,
J. Watson, K. N. Wright and N. Clarke, Bioorg. Med. Chem.
Lett., 2015, 25, 5121–5126.
3
4
4
143.
Foye’s principles of medicinal chemistry, ed. W. O. Foye, 20 J. C. Barrow, P. G. Nantermet, H. G. Selnick, K. L. Glass,
T. L. Lemke and D. A. Williams, Wolters Kluwer Health/
Lippincott Williams & Wilkins, Philadelphia, 7th edn,
P. L. Ngo, M. B. Young, J. M. Pellicore, M. J. Breslin,
J. H. Hutchinson, R. M. Freidinger, C. Condra,
J. Karczewski, R. A. Bednar, S. L. Gaul, A. Stern, R. Gould
and T. M. Connolly, Bioorg. Med. Chem. Lett., 2001, 11,
2691–2696.
2
013.
5
6
M. G. Matera, C. P. Page, L. Calzetta, P. Rogliani and
M. Cazzola, Pharmacol. Rev., 2020, 72, 218–252.
P. Gupta and N. Mahajan, New J. Chem., 2018, 42, 12296– 21 C. Lu, Z. Luo, L. Huang and X. Li, Tetrahedron: Asymmetry,
1
2327.
2011, 22, 722–727.
7
8
9
C. Schneider, Synthesis, 2006, 3919–3944.
R.-E. Parker and N. S. Isaacs, Chem. Rev., 1959, 59, 737–799.
J. Liu, D. Zhou, X. Jia, L. Huang, X. Li and A. S. C. Chan, 23 J. Wu, Y. Tian, S. Wang, M. Pistolozzi, Y. Jin, T. Zhou,
22 Q. Ye, T. Cheng, Y. Zhao, J. Zhao, R. Jin and G. Liu,
ChemCatChem, 2015, 7, 1801–1805.
Tetrahedron: Asymmetry, 2008, 19, 1824–1828.
0 F. Baur, D. Beattie, D. Beer, D. Bentley, M. Bradley, I. Bruce,
G. Roy, L. Xu and W. Tan, Eur. J. Med. Chem., 2017, 126,
61–71.
1
S. J. Charlton, B. Cuenoud, R. Ernst, R. A. Fairhurst, 24 A. González-Antuña, I. Lavandera, P. Rodríguez-González,
B. Faller, D. Farr, T. Keller, J. R. Fozard, J. Fullerton,
S. Garman, J. Hatto, C. Hayden, H. He, C. Howes, D. Janus,
J. Rodríguez, J. I. García Alonso and V. Gotor, Tetrahedron,
2011, 67, 5577–5581.
Z. Jiang, C. Lewis, F. Loeuillet-Ritzler, H. Moser, J. Reilly, 25 M. Nery, M. Azevedo, J. Cardoso, G. Slana, R. Lopes and
A. Steward, D. Sykes, L. Tedaldi, A. Trifilieff, M. Tweed, C. Lopes, Synthesis, 2007, 1471–1474.
S. Watson, E. Wissler and D. Wyss, J. Med. Chem., 2010, 53, 26 M. Nery, M. Azevedo, J. Cardoso, G. Slana, R. Lopes and
675–3684. C. Lopes, Synthesis, 2016, 462–462.
1 A. Bosak, I. Gazić Smilović, G. Šinko, V. Vinković and 27 L.-L. Gan, M.-W. Wang, M.-S. Cheng and L. Pan, Biol.
Z. Kovarik, J. Med. Chem., 2012, 55, 6716–6723. Pharm. Bull., 2003, 26, 323–328.
2 N. Arnold, D. Beattie, M. Bradley, A. Brearley, L. Brown, 28 X. Ge, A. Y.-H. Woo, G. Xing, Y. Lu, Y. Mo, Y. Zhao, Y. Lan,
3
1
1
S. J. Charlton, R. A. Fairhurst, D. Farr, J. Fozard,
J. Fullerton, M. Gosling, J. Hatto, D. Janus, D. Jones,
J. Li, H. Yan, L. Pan, Y. Zhang, B. Lin and M. Cheng,
Eur. J. Med. Chem., 2018, 152, 424–435.
L. Jordan, C. Lewis, J. Maas, C. McCarthy, M. Mercer, 29 X. Ge, Y. Mo, G. Xing, L. Ji, H. Zhao, J. Chen, B. He,
H. Oakman, N. Press, R. Profit, F. Schuerch, D. Sykes,
R. J. Taylor, A. Trifilieff and A. Tuffnell, Bioorg. Med. Chem.
Lett., 2014, 24, 4341–4347.
X. Chen, R. Xing, X. Li, Y. Zhao, J. Li, H. Yan, A. Y.-H. Woo,
Y. Zhang, B. Lin, L. Pan and M. Cheng, Bioorg. Chem.,
2018, 79, 155–162.
This journal is © The Royal Society of Chemistry 2020
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
3
3
0 A. Bosak, A. Knežević, I. Gazić Smilović, G. Šinko and 48 A. V. Marenich, C. J. Cramer and D. G. Truhlar, J. Phys.
Z. Kovarik, J. Enzyme Inhib. Med. Chem., 2017, 32, 789–797. Chem. B, 2009, 113, 6378–6396.
1 D. I. Ball, R. T. Brittain, R. A. Coleman, L. H. Denyer, 49 S. Grimme, J. Chem. Phys., 2006, 124, 034108.
D. Jack, M. Johnson, L. H. C. Lunts, A. T. Nials, 50 S. Grimme, S. Ehrlich and L. Goerigk, J. Comput. Chem.,
K. E. Sheldrick and I. F. Skidmore, Br. J. Pharmacol., 1991,
04, 665–671.
2 J. M. Grisar, G. P. Claxton, T. M. Bare, R. C. Dage,
2011, 32, 1456–1465.
51 L. Goerigk and S. Grimme, J. Chem. Theory Comput., 2011,
7, 291–309.
1
3
H. C. Cheng and J. K. Woodward, J. Med. Chem., 1981, 24, 52 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
27–336.
3
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,
G. A. Petersson, H. Nakatsuji, X. Li, M. Caricato,
A. V. Marenich, J. Bloino, B. G. Janesko, R. Gomperts,
B. Mennucci, H. P. Hratchian, J. V. Ortiz, A. F. Izmaylov,
J. L. Sonnenberg, D. Williams-Young, F. Ding, F. Lipparini,
F. Egidi, J. Goings, B. Peng, A. Petrone, T. Henderson,
D. Ranasinghe, V. G. Zakrzewski, J. Gao, N. Rega, G. Zheng,
W. Liang, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, T. Vreven, K. Throssell, J. A. Montgomery Jr.,
J. E. Peralta, F. Ogliaro, M. J. Bearpark, J. J. Heyd,
E. N. Brothers, K. N. Kudin, V. N. Staroverov, T. A. Keith,
R. Kobayashi, J. Normand, K. Raghavachari, A. P. Rendell,
J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi,
J. M. Millam, M. Klene, C. Adamo, R. Cammi,
J. W. Ochterski, R. L. Martin, K. Morokuma, O. Farkas,
J. B. Foresman and D. J. Fox, Gaussian 16, Gaussian, Inc.,
Wallingford CT, 2016.
3
3
3
3
3
3
3
4
4
4
4
4
3 D. T. Collin, D. Hartley, D. Jack, L. H. Lunts, J. C. Press,
A. C. Ritchie and P. Toon, J. Med. Chem., 1970, 13, 674–680.
4 Y. Rong and A. E. Ruoho, Synth. Commun., 1999, 29, 2155–
2
162.
5 Z.-L. Guo, Y.-Q. Deng, S. Zhong and G. Lu, Tetrahedron:
Asymmetry, 2011, 22, 1395–1399.
6 R. Hett, R. Stare and P. Helquist, Tetrahedron Lett., 1994,
3
5, 9375–9378.
7 M. M. Toteva and J. P. Richard, in Advances in Physical
Organic Chemistry, Elsevier, 2011, vol. 45, pp. 39–91.
8 W. Li, X. Xu, P. Zhang and P. Li, Chem. – Asian J., 2018, 13,
2
350–2359.
9 D. Richter, N. Hampel, T. Singer, A. R. Ofial and H. Mayr,
Eur. J. Org. Chem., 2009, 3203–3211.
0 M. Kazim, M. A. Siegler and T. Lectka, Org. Lett., 2019, 21,
2
326–2329.
1 Y. You, B.-X. Quan, Z.-H. Wang, J.-Q. Zhao and W.-C. Yuan,
Org. Biomol. Chem., 2020, 18, 4560–4565.
2 M. F. McLaughlin, E. Massolo, T. A. Cope and
J. S. Johnson, Org. Lett., 2019, 21, 6504–6507.
3 V. Simeon-Rudolf and B. Juršić, Period. Biol., 1996, 98, 331–
3
4 P. Masson, M.-T. Froment, C. F. Bartels and O. Lockridge,
Eur. J. Biochem., 1996, 235, 36–48.
5 D. Middlemiss and K. J. Ingold, US20110230555A1, 2011.
6 A. D. Becke, J. Chem. Phys., 1993, 98, 5648–5652.
7 T. H. Dunning, J. Chem. Phys., 1989, 90, 1007–1023.
53 R. T. Evans and M. J. McQueen, CRC Crit. Rev. Clin. Lab.
Sci., 1986, 23, 35–64.
54 G. L. Ellman, K. D. Courtney, V. Andres and
R. M. Featherstone, Biochem. Pharmacol., 1961, 7, 88–95.
55 V. Simeon-Rudolf, G. Šinko, A. Štuglin and E. Reiner, Croat.
Chem. Acta, 2001, 74, 173–182.
56 W. N. Aldridge and E. Reiner, Enzyme inhibitors as sub-
strates: interactions of esterases with esters of organopho-
sphorus and carbamic acids, North-Holland, New York,
1972, vol. 26.
35.
4
4
4
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2020