Enantiopure methoxetamine stereoisomers: chiral resolution, conformational analysis, UV-circular dichroism spectroscopy and electronic circular dichroism

Article abstract of DOI:10.1039/d0nj05192f

The chiral molecule, methoxetamine (MXE), demonstrated promising biological effects in management of treatment-resistant depression patients. To satisfy the need for a method capable of providing gram quantities of each enantiopure stereoisomer of MXE to enable advanced biological studies of enantiomers, a protocol was developed to access gram scale quantities of (R)- and (S)-MXE in the form of pharmaceutically acceptable HCl salts employingl-(?)-DTTA andd-(+)-DTTA ((?)-O,O′-di-p-toluoyl-l-tartaric acid and (+)-O,O′-di-p-toluoyl-d-tartaric acid, respectively) as two chiral resolving agents. In contrast to ketamine, the measured specific optical rotation and conformational analysis indicated that the most abundant conformers possess a common axial-methoxyphenyl conformation responsible for the conserved direction of optical rotation in both free base and HCl salt forms of the MXE stereoisomers. Finally, the absolute configuration was unambiguously assigned through matching experimental and calculated ECD spectra. This report offers a gateway to obtain gram scale amounts of enantiopure MXE stereoisomers needed to advance the current knowledge on MXE biology.

Full text of DOI:10.1039/d0nj05192f

NJC
View Article Online
View Journal | View Issue
PAPER
Enantiopure methoxetamine stereoisomers: chiral
resolution, conformational analysis, UV-circular
dichroism spectroscopy and electronic circular
Cite this: New J. Chem., 2021,
45, 4354
dichroism†
bc
Kun Won Lee,‡a Ahmed H. E. Hassan,
‡
Youngdo Jeong,a Seolmin Yoon,a
Seung-Hwan Kim,a Cheol Jung Lee,a Hye Rim Jeon,a Suk Woo Chang,a
Ji-Young Kim,a Dae Sik Jang,a Hee Jin Kim,d Jae Hoon Cheonge and
ac
Yong Sup Lee
*
The chiral molecule, methoxetamine (MXE), demonstrated promising biological effects in management
of treatment-resistant depression patients. To satisfy the need for a method capable of providing gram
quantities of each enantiopure stereoisomer of MXE to enable advanced biological studies of
enantiomers, a protocol was developed to access gram scale quantities of (R)- and (S)-MXE in the form
of pharmaceutically acceptable HCl salts employing L-(ꢀ)-DTTA and D-(+)-DTTA ((ꢀ)-O,O0-di-p-toluoyl-
L-tartaric acid and (+)-O,O0-di-p-toluoyl-D-tartaric acid, respectively) as two chiral resolving agents. In
contrast to ketamine, the measured specific optical rotation and conformational analysis indicated that
the most abundant conformers possess a common axial-methoxyphenyl conformation responsible for
the conserved direction of optical rotation in both free base and HCl salt forms of the MXE
stereoisomers. Finally, the absolute configuration was unambiguously assigned through matching
experimental and calculated ECD spectra. This report offers a gateway to obtain gram scale amounts of
enantiopure MXE stereoisomers needed to advance the current knowledge on MXE biology.
Received 23rd October 2020,
Accepted 21st January 2021
DOI: 10.1039/d0nj05192f
rsc.li/njc
depression and the involvement of multiple molecular targets,
the discovery and development of antidepressants has never
Introduction
Depression is the most common mental illness that impairs been a straightforward task. The inception of the monoamine
psychosocial functioning and quality of life of hundreds of hypothesis in the second half of the 20th century provided a
millions of individuals worldwide. In fact, depression is a gateway for developing several effective antidepressant
complex heterogeneous disease which is provoked by inter- agents acting through the monoamine-based mechanism.
twined diverse genetic, epigenetic, developmental and environmental Monoaminergic-based antidepressants, such as selective
factors.1,2 The World Health Organization anticipates the bur- serotonin reuptake inhibitors (SSRIs) and serotonin norepinephrine
den of major depressive disorder (MDD) will rank first by 2030.3 reuptake inhibitors (SNRIs), have been the only available tools
Complicated by the partial comprehension of neurobiology of to manage depressive disorders for decades.4 Nevertheless, a
significant number of patients are non-responsive to
a Department of Life and Nanopharmaceutical Sciences, Kyung Hee University,
monoaminergic-based antidepressants and, thus, suffer from
treatment-resistant depression (TRD, also known as treatment-
refractory depression). Evidently, the monoamine hypothesis is
not enough to fully explain depression. In addition, the full
efficacy of monoaminergic-based antidepressants is achieved
only after several weeks. Moreover, several patients show
only partial responses. There is a need to consider other
mechanism-based targets to develop newer, faster and safer
antidepressant agents.
Seoul 02447, Korea. E-mail: kyslee@khu.ac.kr
b Department of Medicinal Chemistry, Faculty of Pharmacy, Mansoura University,
Mansoura 35516, Egypt
c Medicinal Chemistry Laboratory, Department of Pharmacy, College of Pharmacy,
Kyung Hee University, Seoul 02447, Korea
d Uimyung Research Institute for Neuroscience, College of Pharmacy,
Sahmyook University, Seoul, Republic of Korea
e School of Pharmacy, Jeonbuk National University, 567 Baekje-daero, Deokjin-gu,
Jeonju-si, Jeollabuk-do 54896, Republic of Korea
† Electronic supplementary information (ESI) available: HRMS and NMR spectra
of the compounds. See DOI: 10.1039/d0nj05192f
Accumulated evidence has confirmed key roles of glutamate
and GABA pathways in depressive disorders, which has opened a
gateway to develop a new class of rapid-acting glutamatergic-based
‡ Both Ahmed H. E. Hassan and Kun Won Lee contributed equally to this work
and should be first authors.
4354 | New J. Chem., 2021, 45, 4354ꢀ4364
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021

View Article Online
Paper
NJC
depressive disorder (MDD) with acute suicidal ideation or
behaviour. Nevertheless, significant drawbacks are associated
with KET and esketamine including dissociative symptoms,
hypertension and other side effects that last for a few hours
post-dose administration. Consequently, esketamine was
FDA-approved under
a ‘‘Risk Evaluation and Mitigation
Strategy (REMS)’’ that necessitates administration of the drug
only in licensed clinics or hospitals. In fact, KET is poorly
selective as indicated by its affinity to m-, k-, and d-opioid
receptors,19 agonistic activity for s1 and s2 receptors,20 antag-
onistic activity for muscarinic receptors,19 and several other
biological targets.21
Methoxetamine (MXE, Fig. 1) might be a better antidepressant
alternative to KET. It possesses a higher NMDA receptor
affinity.22 In addition, it triggers rapid and sustained anti-
depressant effects similar to KET.23 Moreover, it shows better
Fig. 1 Chemical structures of racemic and enantiopure forms of keta-
mine (KET) and methoxetamine (MXE).
antidepressants.5,6 Basically, three types of ionotropic glutamate selectivity as indicated by the absence of negligible affinities to
receptors have been identified which are NMDA receptors, AMPA several off-targets including m-, k-, and d-opioid receptors, s1 and
receptors and kainate receptors (KARs). The extrasynaptic s2 receptors. However, MXE is a racemic mixture of (R)- and
NMDA receptors inhibit long-term potentiation (LTP) and (S)-enantiomers. To date, the eutomer and distomer enantiomers
produce long-term depression (LTD) affecting synaptic plasticity.7 of MXE have not been identified. The development of a suitable
Accordingly, NMDA receptor antagonists might be promising method for the preparation and enantiomeric resolution of gram-
antidepressant therapeutics which is supported by the scales of enantiopure MXE stereoisomers is needed to satisfy the
reduction of immobility in the forced swim (FST) and tail demands of biological studies and identify the eutomer and the
suspension (TST) tests.8 Meanwhile, AMPA receptors’ activation distomer. This might assist in maximizing the therapeutic effects
triggers antidepressant effects and, consequently, AMPA receptor and minimizing the burden. To meet these needs, we report our
activators might serve as useful antidepressant agents.9–11
efforts towards synthesis, enantioseparation and absolute
Recently, ketamine (KET, Fig. 1) was reported as an effective configuration assignment for the free base forms and the
antidepressant for TRD.12–14 It exhibits rapid and long lasting pharmaceutical acceptable HCl salt forms of MXE stereoisomers.
antidepressant effects mainly through glutamatergic-based-
mechanisms, specifically NMDA receptor antagonism and
AMPA receptor activation. Studies indicated that KET exerts a
rapid-onset and strong effects on TRD and bipolar depression,
in addition to minimizing suicidal thoughts. However, KET is a
Results and discussion
Synthesis of racemic (rac)-MXE HCl salt
racemic mixture of (R)- and (S)-enantiomers known as esketamine (rac)-MXE was synthesized via modification of the method
and arketamine, respectively (Fig. 1). A great concern is that reported by Jurasek et al.24 As shown in Scheme 1, 3-methoxy-
enantiomers possibly elicit different biological responses benzonitrile (1) reacted with the Grignard reagent, cyclopentyl-
including therapeutic effects, adverse effects and toxicities.15,16 magnesium bromide, similar to the reported method to afford
Group I of chiral drugs involves chiral drugs whose racemic 3-methoxyphenyl cyclopentyl ketone (2) in an acceptable yield.
mixtures are composed of an eutomer, which is the bioactive Instead of a-bromination employing bromine, as implemented
enantiomer or the enantiomer eliciting the higher pharma- in the reported method, we used copper(II) bromide25 which
cological response, and a distomer which is the less active, afforded a-bromoketone derivative (3) in an excellent yield.
inactive or the enantiomer lacking the desired pharmacological While the reported method isolated the Schiff base formed
response.17 Meanwhile, group II of chiral drugs involves chiral via the reaction with ethylamine, we advanced the crude
drugs whose racemic mixtures have equally bioactive enantiomers. a-hydroxyimino-derivative (4) to the next step. Importantly, we
Regarding KET enantiomers, esketamine was identified as the replaced the poor yielding thermal rearrangement step used in
eutomer exhibiting 5-fold affinity to the NMDA receptor relative the reported method by palladium(II) chloride-catalysed
to the distomer arketamine.18 Accordingly, esketamine has thermal rearrangement26 which significantly increased the
higher efficacy and lower adverse effects and it is capable of isolated yield of (rac)-MXE after recrystallisation as HCl salt
alleviating depression in about half of the patients with TRD ((rac)-MXE HCl) from methanolic hydrochloric acid solution.
within two hours demonstrating an action of duration lasting While the combined yield of Schiff base formation and thermal
for several days to weeks. In addition, repeated dosing results rearrangement steps used in the reported method was around
in a continued relief in responsive patients. In 2019, the FDA 16.7%, the yield over the modified two steps of Schiff base
approved intranasal esketamine (Spravatos) for the treatment formation and palladium(II) chloride-catalysed rearrangement
of adults with TRD and then extended its approval in 2020 to was 49%. 1H NMR, 13C NMR and HRMS data were in agreement
include the treatment of adult patients suffering from major with the chemical structure. The overall isolated yield of
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem., 2021, 45, 4354ꢀ4364 | 4355

View Article Online
NJC
Paper
Scheme 1 Synthesis of (rac)-MXE HCl.
(rac)-MXE HCl after the introduced modifications was 24.2%
which is a clearly much improved yield in comparison to the
4.4% overall yield of (rac)-MXE of the reported method.24
Enantioseparation of (S)-MXE and (S)-MXE
Chiral resolution of (R)- and (S)-MXE stereoisomers. Towards
the development of a cost-effective and practical chiral resolu-
tion protocol capable of yielding gram quantities of the
enantiopure MXE stereoisomers, fractional crystallization of
the diastereomeric salts was attempted. The reported protocol
for enantioseparation of KET stereoisomers using L-(ꢀ)-DTTA,
(ꢀ)-O,O0-di-p-toluoyl-L-tartaric acid,27 resulted in a poor resolution
of MXE stereoisomers, possibly because of the solubility
difference between the diastereomeric salts, which was not
sufficient to afford crystals of acceptable enantiopurity; in
addition to an appreciable solubility of salts in isopropanol
used as the solvent for crystallization. Consequently, the use of
L-(ꢀ)-DTTA as a single resolving agent for MXE stereoisomers
resulted in poor enantiopurity and yield (69% ee and 61% yield
for (ꢀ)-MXE; 32% ee and 32% yield for (+)-MXE). Instead of
employing only L-(ꢀ)-DTTA as a single resolving agent, a new
protocol (Schemes 2 and 3) was developed utilizing two resolving
agents; L-(ꢀ)-DTTA and D-(+)-DTTA; (+)-O,O0-di-p-toluoyl-D-
tartaric acid, for fractional crystallization of the diastereomeric
salts from methanolic solutions of salts formed with each chiral
resolving agent. For each resolved stereoisomer, the developed
protocol involved two sequential fractional crystallization steps
to enrich its enantiopurity. Thus, as shown in Scheme 2, (rac)-
MXE was first converted into L-(ꢀ)-DTTA salt in methanolic
solution which enabled crystallization of (ꢀ)-MXE as a salt of
L-(ꢀ)-DTTA. To enrich the enantiopurity, a second crystallization
step afforded (ꢀ)-MXE L-(ꢀ)-DTTA salt in 51% yield over the two
crystallization steps. To convert to the pharmaceutical acceptable
HCl salt, the free (ꢀ)-MXE base was liberated via treatment with
aqueous sodium bicarbonate followed by extraction with toluene
and then crystallization from methanolic HCl solution to afford
Scheme 2 Established gram scale (R)-MXE HCl enantioseparation
protocol.
Scheme 3 Established gram scale (S)-MXE HCI enantioseparation
protocol.
4356 | New J. Chem., 2021, 45, 4354ꢀ4364
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021

View Article Online
Paper
NJC
the desired (R)-MXE HCl crystals in 91% yield (based on (R)-MXE form might be a reflection of the impact of solvent effects
on specific rotation.30 As will be shown below, the assignment
L-(ꢀ)-DTTA salt).
To obtain the other stereoisomer as an enantiopure HCl salt, of absolute configuration proved that the first isolated
i.e. (S)-MXE HCl, the recovered filtrates from the two crystal- stereoisomer is (R)-(ꢀ)-MXE HCl while the second isolated
lization steps of (R)-MXE L-(ꢀ)-DTTA (Scheme 3) were combined stereoisomer is (S)-(+)-MXE HCl.
and treated with sodium bicarbonate to liberate the free base
Establishment of chiral chromatographic conditions for
form of MXE. Using the second chiral resolving agent, it enantioseparation of free and salt forms of (R)- and (S)-MXE.
was converted into D-(+)-DTTA salt which yielded (S)-MXE The development of a chromatographic method for chiral
D-(+)-DTTA crystals upon crystallization from methanol. Again, resolution of the various forms of free base and salts of
a second crystallization step towards enrichment of enantiopurity (R)- and (S)-MXE enantiomers can serve both preparative and
afforded the desired (S)-MXE D-(+)-DTTA crystals in 57% yield analytical purposes. At an analytical scale, it can be a suitable
over two crystallization steps. Herein also, the liberation of the tool to assess the enantiopurity of the resolved enantiomers.
free base form using aqueous sodium bicarbonate and extraction In addition, it can be extrapolated to large-scale preparative
by toluene followed by crystallization from methanolic HCl enantioseparation towards obtaining an enantiopure drug.
solution afforded the pharmaceutical acceptable form (S)-MXE However, such chromatographic conditions have not been
HCl crystals in 91% yield (based on (S)-MXE D-(+)-DTTA salt).
established yet for MXE. A chiral HPLC condition was reported
Employing our established chiral HPLC conditions shown for KET using a CHIRALPAKs IA column, which is an amylose-
below, the excellent enantiopurities of the obtained crystals based chiral stationary phase bearing spatially organized 3,5-
were confirmed (98% ee and 99% ee for the first and the second dimethylphenylcarbamates on the stereochemically oriented
isolated stereoisomers, respectively; Fig. S6, ESI†). The hydroxyl groups, and n-hexane/dichloromethane/diethylamine
measurement of optical properties demonstrated that the first (75/25/0.1) as a mobile phase.31 Having the more polar methoxy
isolated stereoisomer demonstrated levorotatory properties in substituent at the 3-position instead of non-polar chloro sub-
all of the assessed free base and salt forms (specific optical stituent at the 2-position of KET, the molecular structure of
rotation values of ꢀ11.11, ꢀ4.81, ꢀ15.11 and ꢀ1701 for the free MXE is distinct from KET. Consequently, significant affinity
base form in DMF solution, L-(ꢀ)-DTTA salt form in DMF, and changes of MXE enantiomers towards stationary and mobile
HCl salt form in ethanolic and aqueous solutions, respectively; phases might arise. In addition, the positively charged nitrogen
Table 1). Meanwhile, the second isolated stereoisomer was atom in the case of MXE salts might be another factor that can
confirmed to exhibit dextrorotatory properties in all of the induce affinity differences relative to KET as the reported
assessed free base and salt forms (specific optical rotation method was developed for KET’s uncharged free base form.
values of +11.81, +6.21, +14.71 and +1651 for free base form in Moreover, the dichloromethane-based mobile phase used in
DMF solution, L-(ꢀ)-DTTA salt form in DMF, and HCl salt form the reported method employed for KET bears several inherent
in ethanolic and aqueous solutions, respectively; Table 1). This disadvantages including its limited UV transmission levels
maintenance of the optical rotation direction is a notable (transmittance of only 5% at 230 nm) rendering it incompatible
behaviour distinguished from the previously reported inversion with diode-array detectors.32 Therefore, we started to develop
of optical rotation direction that has been observed for KET free new chiral HPLC conditions for enantioseparation of the free
base and HCl forms. The molecular basis for the observed base and salt forms of (R)- and (S)-MXE enantiomers. First, we
optical rotation inversion in the case of KET stemmed from started to establish chiral chromatographic conditions for
the fact that the equatorial-chlorophenyl conformation is the enantioseparation of the free base form of MXE. Eight
most-stable conformer of the free base form while the axial- amylose/cellulose-based chiral stationary phases bearing structurally
chlorophenyl conformation is the most-stable conformer of the different carbamates spatially organized on the stereochemically
HCl salt form.28,29 It might be inferred from the obtained oriented hydroxyl groups (CHIRALPAKs IA-3, IB-3, IC-3, ID-3,
results herein that a similar most stable conformer might be IE-3, IF-3, IG-3 and IH-3) were screened using two different
shared between the free base and the corresponding salt forms mobile phase systems (acetonitrile/Dist. water (60/40) and
investigated of MXE. In addition, the large difference in specific ethanol/diethylamine (100/0.1)). Configurationally, amylose is
rotation between aqueous and ethanolic solutions of the salt poly(a-D-glucopyranoside) while cellulose has the inverted
Table 1 Overall yields, enantiomeric excess and specific optical rotation values of the resolved MXE HCl stereoisomers
(ꢀ)-MXE L-(ꢀ)-DTTA
1.57 g, 51%
(+)-MXE D-(+)-DTTA
(ꢀ)-MXE Free base
—
—
—
(+)-MXE Free base
(ꢀ)-MXE HCl
0.64 g, 46%
98%
(+)-MXE HCl
Yielda
1.76 g, 57%
—
+6.21
—
—
—
—
+11.81
—
0.72 g, 52%
99%
—
+14.71
+1651
eeb
—
[a]2D5 (c 1.0, DMF)
[a]2D5 (c 1.0, EtOH)
[a]2D5 (c 0.1, H2O)
ꢀ4.81
—
—
ꢀ11.11
ꢀ15.11
—
—
—
ꢀ1701
a
Yield was calculated based on the free base form of (rac)-MXE employed for enantioseparation assuming 100% of each enantiomer.
ee: enantiomeric excess.
b
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem., 2021, 45, 4354ꢀ4364 | 4357

View Article Online
NJC
Paper
anomeric stereochemical configuration; i.e. poly(b-D-gluco- (50/50/0.1) as a mobile phase employing CHIRALPAKs IG-3, the
pyranoside). The employed stationary phases CHIRALPAKs same stationary phase that successfully resolved the free base
IB-3 and IC-3 were based on cellulose bearing 3,5-dimethyl- form. Unfortunately, (R)- and (S)-enantiomers were not resolved,
phenylcarbamate or 3,5-dichlorophenylcarbamate, respectively, possibly because of induced alteration of affinity requirements
spatially organized on the stereochemically oriented hydroxyl by positively charged nitrogen of the salt form or the change of
groups (Fig. S1, ESI†). Meanwhile, stationary phases CHIRALPAKs mobile phase composition. Accordingly, five chiral stationary
IA-3, ID-3, IE-3, IF-3, IG-3 and IH-3 were based on amylose bearing phases bearing phenylcarbamates having different chloro and/or
3,5-dimethylphenylcarbamate, 3-chlorophenylcarbamate, 3,5-di- methyl-substitution patterns spatially organized on the stereo-
chlorophenylcarbamate, 3-chloro-4-methylphenylcarbamate, 3- chemically oriented hydroxyl groups of amylose/cellulose (CHIR-
chloro-5-methylphenylcarbamate or (S)-a-methylbenzylcarbamate, ALPAKs IA-3, IB-3, ID-3, IE-3 and IF-3) were evaluated employing
respectively, spatially organized on the stereochemically oriented ethanol/methanol/diethylamine (50/50/0.1) as a mobile phase.
hydroxyl groups (Fig. S1, ESI†). The variation in the stereochemical Four stationary phases with either chloro or methyl phenylcarba-
configuration of the stationary phase combined with the diversely mate (CHIRALPAKs IA-3, IB-3, ID-3 and IE-3) could not resolve
substituted carbamates spatially organized on the stereochemically the HCl salt form of MXE enantiomers. Meanwhile, satisfactory
oriented hydroxyl groups of these stationary phases is anticipated enantioseparation was achieved only using the stationary phase
to demonstrate different affinities towards enantiomers enabling bearing spatially organized 3-chloro-4-methylphenylcarbamate
differentiation of MXE (R)- and (S)-enantiomers. We observed that on the stereochemically oriented hydroxyl groups of amylose; i.e.
CHIRALPAKs IA-3, the column reported to resolve the free base CHIRALPAKs IF-3 (Fig. S3, ESI†). Next, we checked whether the
form of (R)- and (S)-KET enantiomers, failed to resolve the free shift from 3-chloro-5-methylphenylcarbamate (CHIRALPAKs
base form of (R)- and (S)-MXE enantiomers upon using both of the IG-3) to 3-chloro-4-methyl (CHIRALPAKs IF-3) was because of
employed mobile phase systems. Switching to CHIRALPAKs IB-3, the change in the composition of the mobile phase or the
a cellulose-based stationary phase, i.e. the inverted anomeric presence of a positively charged nitrogen in the salt form. In
stereochemical configuration, while maintaining the same carba- this regard, enantioseparation of the free base form of (rac)-
mate residue of CHIRALPAKs IA-3, failed also to resolve free base MXE was evaluated using CHIRALPAKs IF-3 as a stationary
forms of (R)- and (S)-MXE enantiomers. The use of any of the two phase and ethanol/methanol/diethylamine (50/50/0.1) as a
employed mobile phase systems while replacing the 3,5- mobile phase. The free base form of (rac)-MXE was success-
dimethylphenylcarbamates of CHIRALPAKs IA-3 and IB-3 by 3,5- fully resolved under these conditions (Fig. S4, ESI†) indicating
dichlorophenylcarbamate failed also to resolve the free base form that the shift from the 3-chloro-5-methyl substitution pattern
of (R)- and (S)-MXE enantiomers regardless of being spatially of the phenylcarbamate (CHIRALPAKs IG-3) to the 3-chloro-4-
organized on the stereochemically oriented hydroxyl groups of methyl substitution pattern of the phenylcarbamate (CHIR-
a- or b-anomeric configurations, i.e. CHIRALPAKs IC-3 or IE-3. ALPAKs IF-3) might be attributed mainly to the change in
Accordingly, we focused on the stationary phases based on the composition of the mobile phase. It also suggests the
a-anomeric configuration; i.e. amylose, bearing diverse carbamate latter condition as a suitable condition for enantioseparation
residues, 3-chlorophenyl, 3-chloro-4-methylphenyl, 3-chloro-5- of both the free base and HCl salt forms of MXE. The
methyl or (S)-a-methylbenzyl derivatives of carbamate moiety universality of these optimized conditions to other salt forms
(CHIRALPAKs ID-3, IF-3, IG-3 or IH-3). None of them could resolve of MXE was checked using enantiopure MXE salts formed
(R)- and (S)-MXE enantiomers when the employed mobile phase with both the L-(ꢀ)-DTTA and D-(+)-DTTA. As indicated in
was acetonitrile/dist. water (60/40). Only 3-chloro-5-methyl- Fig. S5 (ESI†), significantly different retention times were
phenylcarbamate spatially organized on the stereochemically observed for each of them. Collectively, these results
oriented hydroxyl groups of amylose, i.e. CHIRALPAKs IG-3 could suggest that the established optimized conditions employing
resolve successfully the free base forms of (R)- and (S)-MXE chiral stationary phase CHIRALPAKs IF-3, which is composed
enantiomers upon using ethanol/diethylamine (100/0.1) as a of spatially organized 3-chloro-4-methylphenylcarbamate on
mobile phase (Fig. S2, ESI†). Meanwhile the other employed the stereochemically oriented hydroxyl groups of amylose,
stationary phases having other carbamates spatially organized on in combination with ethanol/methanol/diethylamine (50/50/0.1)
the stereochemically oriented hydroxyl groups failed to resolve the as the mobile phase, might serve as universal conditions
free base form of (R)- and (S)-MXE enantiomers upon using for chiral chromatographic separation of different forms
ethanol/diethylamine (100/0.1) as a mobile phase.
of MXE.
Conformational analysis of free base and HCl forms of MXE
Pharmaceutically, salt forms are, in general, more preferable
than the free base forms. This might be attributed, at least in stereoisomers. According to literature reports, two different
part, to their higher stability and better aqueous solubility. In conformations for KET and KET HCl are responsible for the
fact, MXE HCl salt might be the practically used pharmaceutical observed inversion of optical rotation direction. A free KET base
form. Therefore, we addressed the establishment of chiral chro- form exists in the equatorial-chlorophenyl conformation while
matographic conditions for resolution of its enantiomers. Based on KET HCl exists in the axial-chlorophenyl conformation
our gained knowledge regarding chromatographic enantiosepara- (Fig. 2A).28,29 In contrast, the found retainability of the
tion of free base forms of (rac)-MXE, we tested the resolution of direction of optical rotation of MXE enantiomers in both of
the (rac)-MXE HCl salt form using ethanol/methanol/diethylamine the free base and its corresponding HCl salt form suggests the
4358 | New J. Chem., 2021, 45, 4354ꢀ4364
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021

View Article Online
Paper
NJC
phenomenon, conformational analysis of free base and HCl
salt forms of (R)- and (S)-MXE was conducted. In this regard in
silico calculations employing the systematic conformational
search algorithm were conducted using the Merk molecular
force field (MMFF). The results showed that an ensemble of
11 conformers represents almost 95% of the conformer
populations of the free base form of (R)- and (S)-MXE while
95% of the conformer populations are represented by only 4
conformers in the case of the HCl salt form of (R)- and (S)-MXE.
As shown in Table 2, only 2 conformers among the 4 most
abundant conformers of the free base form of (R)- and (S)-MXE
represented more than 50% of the total conformers’ population
(populated according to a Boltzmann distribution). As shown in
Fig. 2B and C, both of these two conformers for both of (R)- and
(S)-MXE adopt an axial-methoxyphenyl conformation. In case
of HCl forms of (R)- and (S)-MXE, only two conformers repre-
sented more than 92% of the total population of conformers
(Table 2) and, notably, they were significantly energetically
more stable than other conformers. Similar to the most two
abundant conformers of the free base form, the conformation
of these two conformers showed an axial-methoxyphenyl
configuration (Fig. 2D and E). These results are consistent with
the inferred existence of a shared common conformation
between the free base and the HCl salt form responsible for
the retainability of the direction of optical rotation which
contrasts the inversion of the direction of optical rotation in
Table 2 Relative energies of the four most abundant conformers among the
generated conformers populated according to the Boltzmann distribution
Cumulative
Boltzmann Boltzmann
Relative
Stereoisomer Conform. energy (kJ molꢀ1
)
weights
weights
1
2
3
4
0.00
0.35
2.62
2.80
0.278
0.241
0.096
0.090
0.278
0.518
0.615
0.704
1
2
3
4
0.00
0.35
2.62
2.80
0.278
0.241
0.097
0.090
0.278
0.519
0.616
0.706
1
2
3
4
0.00
0.58
8.44
8.60
0.516
0.409
0.017
0.016
0.516
0.925
0.942
0.959
Fig. 2 Most stable conformers of stereoisomers of KET and MXE: (a) reported
conformations of free base and HCl salt of (S)-KET; (b) calculated most stable
conformers of (R)-MXE free base form; (c) calculated most stable conformers
of (R)-MXE HCl form; (d) calculated most stable conformers of (S)-MXE free
base form; (e) calculated most stable conformers of (S)-MXE HCl form.
1
2
3
4
0.00
0.58
8.44
8.60
0.516
0.409
0.017
0.016
0.516
0.925
0.942
0.959
existence of a common most stable conformer. To check this
suggested explanation as
a reasonable basis for this
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem., 2021, 45, 4354ꢀ4364 | 4359

View Article Online
NJC
Paper
the case of free base form of KET’s stereoisomers relative to was consistent with the pattern of the calculated ECD spectrum
their corresponding HCl salts. using the CPCM solvent model for the S model, suggesting the
Assignment of absolute configurations of free base and HCl absolute configuration of (ꢀ)-MXE HCl as S. In comparison to
forms of MXE enantiomers. Towards assigning the absolute the CPCM model, the ECD spectra calculated using the SMD
configuration of the isolated MXE enantiomers as free bases solvent model for both R and S models of MXE HCl salts did
and HCl forms, experimental circular dichroism (CD) spectra of not show the pronounced CE observed at 205 nm in the
the separated enantiomers were acquired and compared to experimental CD spectra (Fig. S7, ESI†).
the calculated electronic circular dichroism (ECD) spectra of
In summary, the absolute configurations were unambiguously
(R)- and (S)-MXE enantiomers as free bases and HCl salts. assigned based on matching experimental CD spectra with ECD
Because of the high cost of ab initio calculations, there is a spectra calculated by employing the CPCM solvent model. Mean-
need to implement calculation methods with acceptable cost- while, the ECD spectra calculated using the SMD solvent model
accuracy compromise. Accordingly, we have employed the were less reliable for assignment of absolute configuration of MXE
sufficiently accurate and cost-effective time-dependent density enantiomers. In addition, the calculations employing the SMD
functional theory (TDDFT) method.33 Selecting a suitable solvent model for MXE HCl salts were of much lower accuracy
combination of functional/basis set is important for TDDFT relative to calculations of MXE free bases. Collectively, these
calculations. It is noteworthy that benchmarking studies have results confirm that there is no inversion of the direction of
shown that the performance of range-separated hybrid functions optical rotation between the free base and HCl salt forms of MXE
such as CAM-B3LYP is better than the just hybrid functions such which is distinct from the case of KET.
as B3LYP for the calculation of ECD spectra.34 In addition, using
Ahlrichs basis sets such as SVP for TDDFT calculations is more
reliable than the split-valence Pople basis sets such as 6-31+G.
Accordingly, ECD calculations were performed using
CAM-B3LYP/SVP as the functional/basis set. Two solvent models
Experimental
Chemistry
were employed in parallel to run separate calculations of
General. All commercially available reagents and solvents
ECD spectra. The first solvent model was the ‘‘conductor-like were purchased and used without further purification. All
polarizable continuum model’’ (CPCM; calculated ECD spectra reactions were monitored by thin-layer chromatography using
are shown in Fig. 3) while the second was the ‘‘solvation model E. Merck silica gel 60 F254 precoated plates (0.25 mm; Darm-
based on density’’ (SMD; calculated ECD spectra are shown in stadt, Germany). The TLC spots were visualized under a UV
Fig. S7, ESI†).
lamp or using staining solutions such as p-anisaldehyde
As shown in Fig. 3, the experimental ECD spectrum of the solution and ninhydrin solution. 1H and 13C NMR spectra were
(ꢀ)-MXE free base form displayed a positive cotton effect (CE) at recorded on a Bruker Avance 400 spectrometer (400 MHz) or a
203 nm (De +10.3) and negative CEs at 227 nm (De ꢀ2.8) and JEOL JNM-ECZ500R spectrometer (500 MHz). Chemical shifts
297 nm (De ꢀ2.4). This experimental ECD spectrum (Fig. 3) was (d) are reported in ppm relative to tetramethylsilane (0 ppm).
in agreement with the pattern of the calculated ECD spectrum High resolution mass spectra (HRMS) were recorded on a Jeol
using the CPCM solvent model for the R model of MXE free AccuTOF (JMS-T100TD) equipped with a DART (direct analysis
base, suggesting the absolute configuration of the (ꢀ)-MXE in real time) ion source from ionsense, Tokyo, Japan in the
base form as R. Meanwhile, the experimental ECD spectrum of positive modes. Optical rotations were measured by using a
the (+)-MXE free base form displayed a negative CE at 205 nm Jasco P-2000 polarimeter (light source = Na, 589 nm; Hacketts-
(De ꢀ11.3) and positive CEs at 227 nm (De +1.8) and 300 nm town, USA). Enantiomeric purity assay was carried out by using
(De +1.7). This experimental ECD spectrum (Fig. 3) was in an optimized chiral HPLC using Agilent 1100 Series Capillary
agreement with the pattern of the calculated ECD spectrum LC (Waldbronn, Germany) using Chiralpaks (3 mm particle
using the CPCM solvent model for the S model, suggesting the size, 4.6 mm ꢁ 150 mm; Tokyo, Japan) as a stationary phase
absolute configuration of the (+)-MXE base as S. Regarding the and ethanol/methanol/diethylamine = 50 : 50 : 0.1 as a mobile
(ꢀ)-MXE HCl form, there was a positive CE at 205 nm (De +11.9) phase in isocratic gradient runs at 0.4 mL minꢀ1. Electronic
and a negative CE at 294 nm (De ꢀ2.9). In comparison to the circular dichroism (ECD) spectra were recorded on a JASCO
CPCM model, ECD spectra calculated using the SMD solvent J-1100(JASCO, Tokyo, Japan) spectropolarimeter.
model for both R and S models of the MXE free base showed
Preparation of cyclopentyl-3-methoxyphenylketone (2)24. 3-
more amplified CEs near the 227 nm and 300 nm wave lengths Methoxybenzonitrile (1, 4.59 mL, 37.55 mmol) was placed in a
suggesting that calculations using the SMD model were less round bottom flask under nitrogen. After cooling to 0 1C,
accurate relative to the CPCM model (Fig. S7, ESI†).
cyclopentylmagnesium bromide (2 M diethyl ether solution,
In case of HCl salts of MXE enantiomers, the experimental 28.17 mL, 56.33 mmol) was added slowly. The reaction
ECD spectrum (Fig. 3) was consistent with the calculated ECD temperature was allowed to increase to ambient temperature
spectrum using the CPCM solvent model for the R model, and the mixture was stirred for 6 hours. The reaction was
suggesting the absolute configuration of (ꢀ)-MXE HCl as R. quenched with HCl (2 N solution, 5 mL) and stirred at ambient
In addition, the (+)-MXE HCl form showed a negative CE at temperature for 15 minutes. pH was rendered alkaline with
205 nm (De ꢀ10.6) and a positive CE at 294 nm (De +2.3), which NaOH (2 N solution) then the mixture was extracted with
4360 | New J. Chem., 2021, 45, 4354ꢀ4364
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021

View Article Online
Paper
NJC
Fig. 3 Comparison of experimental and calculated electronic circular dichroism (ECD) spectra of isolated stereoisomers of MXE free base and their HCl
forms using the conductor-like polarizable continuum model (CPCM).
diethyl ether and H2O. After drying (anhydrous MgSO4) the bromide (8.53 g, 38.18 mmol). The mixture was heated and
extract was concentrated under reduced pressure, and purified stirred under reflux conditions for 4 hours. After cooling
by column chromatography (EtOAc/n-hexane = 1 : 40) to afford to ambient temperature, the reaction mixture was filtered
compound 2 (3.90 g, 19.09 mmol, 51%).
and concentrated. The concentrated filtrate was suspended in
1H NMR (400 MHz, CDCl3) d 7.53 (d, J = 7.6 Hz, 1H), 7.49 (t, hexane, filtered again and then concentrated to obtain com-
J = 2.2 Hz, 1H), 7.34 (t, J = 8.0 Hz, 1H), 7.01 (dd, J = 8.2, 2.3 Hz, pound 3 (5.26 g, 18.58 mmol, 97%) as a crude yellow
1H), 3.82 (s, 3H), 3.67 (p, J = 7.9 Hz, 1H), 1.93–1.84 (m, 4H), transparent oil.
1.73–1.57 (m, 4H).
1H NMR (400 MHz, CDCl3) d 7.78 (d, J = 7.6 Hz, 1H), 7.65 (t,
Preparation of 1-bromocyclopentyl-3-methoxyphenylketone J = 2.2 Hz, 1H), 7.35 (t, J = 7.9 Hz, 1H), 7.09 (dd, J = 8.2, 1.9 Hz,
(3)24. Compound 2 (3.90 g, 19.09 mmol) was added to a round 1H), 3.85 (s, 3H), 2.55–2.38 (m, 4H), 2.11–2.00 (m, 2H), 1.85–
bottom flask, followed by ethyl acetate (30 mL), and copper(II) 1.73 (m, 2H).
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem., 2021, 45, 4354ꢀ4364 | 4361

View Article Online
NJC
Paper
Preparation of 2-ethylamino-2-(3-methoxyphenyl)cyclohexan- (ꢀ)-O,O0-di-p-toluoyl-L-tartarate (1.57 g, 2.48 mmol, 51% yield over
1-one ((rac)-MXE) and the corresponding hydrochloride salt two crystallization steps) while the filtrate (filtrate B) was saved for
((rac)-MXE HCl)24. In a sealed tube under a nitrogen atmosphere the separation of the other stereoisomer after combination with
and cooled to ꢀ40 1C an excess of ethylamine was placed. filtrate A.
Compound 3 (2.00 g, 7.06 mmol) was added slowly. The reaction
White solid; mp: 121–131 1C; [a]2D5: ꢀ4.81 (c 1.0, DMF);
mixture was stirred for 6 hours at ꢀ40 1C, then the temperature 1H NMR (500 MHz, DMSO-d6) d 7.84 (d, J = 8.0 Hz, 4H), 7.40
was allowed to increase slowly to ambient temperature and (t, J = 8.0 Hz, 1H), 7.32 (d, J = 8.6 Hz, 4H), 7.02 (dd, J = 8.6,
stirred for 12 hours at ambient temperature. The reaction was 2.3 Hz, 1H), 6.92–6.89 (m, 2H), 5.65 (s, 2H), 3.76 (s, 3H), 2.98
quenched with water and extracted with diethyl ether. The (d, J = 13.7 Hz, 1H), 2.57–2.52 (m, 1H), 2.36–2.18 (m, 9H), 1.98
extract was concentrated under reduced pressure to afford (td, J = 13.0, 3.2 Hz, 1H), 1.89–1.87 (m, 1H), 1.73 (d, J = 12.0 Hz,
brown oily crude compound 4 (1.75 g) which was dissolved in 1H), 1.62–1.49 (m, 2H), 1.04 (t, J = 7.2 Hz, 3H); 13C NMR
decalin (8 mL) and placed in a sealed tube. The temperature was (125 MHz, DMSO-d6) d 207.00, 168.50, 165.41, 160.31, 144.29,
increased to 190 1C followed by the addition of palladium(II) 130.97, 129.87, 129.76, 127.39, 120.78, 115.27, 114.42, 72.70,
chloride (0.1 g) divided into two portions with a two hour interval 71.21, 55.74, 37.30, 32.97, 27.30, 21.81, 21.71, 12.25.
between additions. After heating for 10 hours, the reaction
Preparation of (R)-(ꢀ)-methoxetamine hydrochloride acid
mixture was cooled, diethyl ether and water were added followed salt. (R)-(ꢀ)-Methoxetamine (ꢀ)-O,O0-di-p-toluoyl-L-tartarate
by 4 N HCl to adjust the pH to 2–3. The mixture was extracted (1.57 g, 2.48 mmol) was treated with NaHCO3 aqueous solution
three times with diethyl ether followed by the addition of 2 N and extracted three times with toluene, dried (anhydrous
NaOH to re-adjust the aqueous phase to pH 10–11 then extracted MgSO4) and evaporated under reduced pressure. The residue
again three times, dried over anhydrous MgSO4, concentrated was treated with hydrochloric acid (1.5 eq. of 1.25 M metha-
under reduced pressure and purified by column chromatogra- nolic solution) and stirred at 0 1C for 30 minutes. After
phy (EtOAc/n-hexane = 1 : 2) to afford the free base form of evaporation under reduced pressure, it was re-dissolved in
methoxetamine (rac)-MXE. The free base was converted into methanol (5 mL) and recrystallized from diethyl ether at 0 1C
the HCl salt through recrystallization from methanolic HCl to afford (R)-(ꢀ)-methoxetamine hydrochloride acid salt (0.64 g,
solution to afford (rac)-MXE HCl (0.99 g, 3.50 mmol, 49% over 91% yield, 98% ee, Chiralpak IF-3, EtOH/MeOH/DEA = 50/50/
2 steps) as a white solid.
0.1, 0.5 mL minꢀ1, rt: 5.75 min).
1
(rac)-MXE: H NMR (500 MHz, CDCl3) d 7.28 (t, J = 7.9 Hz,
White solid; mp: 235–245 1C; [a]2D5: ꢀ15.11 (c 1.0, EtOH,
1H), 6.82 (d, J = 2.9 Hz, 1H), 6.81–6.79 (m, 1H), 6.77 (t, J = hydrochloride salt); [a]2D5: ꢀ1701 (c 0.1, H2O, hydrochloride salt);
1.9 Hz, 1H), 3.88 (s, 3H), 2.88–2.84 (m, 1H), 2.41–2.37 (m, 1H), [a]2D5: ꢀ11.11 (c 0.5, EtOH, free base); H NMR (500 MHz, D2O)
1
2.34–2.26 (m, 2H), 2.09–1.92 (m, 4H), 1.87–1.66 (m, 4H), 0.99 d 7.51 (t, J = 8.0 Hz, 1H), 7.15 (dd, J = 8.6, 2.3 Hz, 1H), 7.01 (d, J =
(t, J = 7.1 Hz, 3H). 13C NMR (125 MHz, CDCl3) d 211.27, 160.08, 8.0 Hz, 1H), 6.95 (s, 1H), 3.84 (s, 3H), 3.20 (d, J = 14.3 Hz, 1H),
141.05, 129.89, 119.45, 113.27, 112.35, 69.79, 55.34, 39.82, 2.87 (dt, J = 19.5, 7.4 Hz, 1H), 2.57–2.48 (m, 3H), 2.04–1.92 (m,
36.67, 36.07, 27.68, 22.44, 15.73.
3H), 1.80–1.67 (m, 2H), 1.14 (t, J = 7.2 Hz, 3H); 13C NMR
(rac)-MXE HCl: 1H NMR (500 MHz, D2O) d 7.51 (t, J = 8.3 Hz, (125 MHz, D2O) d 209.00, 160.08, 131.36, 131.06, 120.81, 115.93,
1H), 7.16 (dt, J = 8.6, 2.3 Hz, 1H), 7.01 (d, J = 8.0 Hz, 1H), 6.96 114.12, 72.03, 55.57, 39.01, 37.14, 32.37, 27.38, 21.19, 10.66.
(s, 1H), 3.84 (s, 3H), 3.20 (dd, J = 13.5, 2.6 Hz, 1H), 2.92–2.83
Enantioseparation of (S)-(+)-methoxetamine (+)-O,O0-di-p-
(m, 1H), 2.58–2.45 (m, 3H), 2.06–1.88 (m, 3H), 1.82–1.67 toluoyl-D-tartarate. The residue after evaporation under reduced
(m, 2H), 1.14 (t, J = 7.2 Hz, 3H); 13C NMR (125 MHz, D2O) d pressure of the combined filtrates A and B was treated with
209.01, 160.26, 131.46, 131.07, 120.81, 115.93, 114.13, 72.04, saturated NaHCO3 (aqueous solution), extracted three times
55.57, 39.01, 37.14, 32.37, 27.38, 21.19, 10.66; HRMS m/z with toluene, dried (anhydrous MgSO4) and re-evaporated
+
calculated for C15H22NO2 [M + H]+ 248.1645, found 248.1671. under reduced pressure to afford the free base form of methox-
etamine as a residue (1.54 g, 6.22 mmol). Methanol (20 mL) and
Chiral resolution of (rac)-MXE into enantiopure stereoisomers
(+)-O,O0-di-p-toluoyl-D-tartaric acid (2.40 g, 6.22 mmol) were
Enantioseparation of (R)-(ꢀ)-methoxetamine (ꢀ)-O,O0-di-p- added to the obtained free base form. The mixture was heated
toluoyl-L-tartarate. (ꢀ)-O,O0-Di-p-toluoyl-L-tartaric acid (3.74 g, under reflux with stirring for 2 hours, then allowed to cool
9.69 mmol) was added to a methanolic solution (30 mL) of (rac)- slowly to ambient temperature and set aside to crystallize for
methoxetamine (2.40 g, 9.69 mmol). The mixture was heated 12 hours at ambient temperature. The formed crystals (2.19 g,
under reflux with stirring for 2 hours, allowed to cool slowly to 3.46 mmol) were collected by filtration. The isolated crystals
ambient temperature and set aside to crystallize for 12 hours at were recrystallized again from methanol (8 mL) after heating
room temperature. The crystallized solid (2.16 g, 3.41 mmol) under reflux with stirring then slow cooling and standing still
was collected by filtration and the filtrate (filtrate A) was saved at ambient temperature for 12 hours. The formed crystals were
for separation of the other stereoisomer. The isolated crystals collected by filtration to afford (S)-(+)-methoxetamine (+)-O,O0-
were recrystallized again from methanol (8 mL) after heating di-p-toluoyl-D-tartarate (1.76 g, 2.77 mmol, 57% yield over two
under reflux with stirring then slowly cooled and stood still crystallization steps).
at ambient temperature for 12 hours. The enantioenriched crys-
White solid; mp: 115.5–125.5 1C; [a]2D5: +6.21 (c 1.0, DMF);
tals were collected by filtration to afford (R)-(ꢀ)-methoxetamine 1H NMR (500 MHz, DMSO-d6) d 7.83 (d, J = 8.6 Hz, 4H), 7.40 (t,
4362 | New J. Chem., 2021, 45, 4354ꢀ4364
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021

View Article Online
Paper
NJC
J = 8.0 Hz, 1H), 7.31 (d, J = 8.0 Hz, 4H), 7.02 (dd, J = 8.6, 2.3 Hz, overall yield). Employing two chiral resolving agents L-(ꢀ)-DTTA
1H), 6.91 (d, J = 8.0 Hz, 1H), 6.88 (d, J = 2.3 Hz, 1H), 5.63 (s, 2H), and D-(+)-DTTA and conducting two sequential fractional
3.76 (s, 3H), 2.98 (d, J = 13.7 Hz, 1H), 2.52–2.57 (m, 1H), 2.16– crystallization steps for each enantiomer afforded in the last
2.39 (m, 9H), 1.88–1.99 (m, 2H), 1.75 (d, J = 11.5 Hz, 1H), 1.50– steps gram scales of pharmaceutical acceptable HCl forms of
1.63 (m, 2H), 1.03 (t, J = 7.2 Hz, 3H); 13C NMR (125 MHz, both (R)- and (S)-MXE enantiomers in excellent ee and reasonable
DMSO-d6) d 207.51, 168.42, 165.36, 160.26, 144.28, 130.87, yields. In contrast to the reported inversion of optical rotation
129.84, 127.36, 120.69, 114.95, 114.38, 72.31, 71.10, 55.74, direction of KET HCl salts relative to the free base form, the
37.26, 33.62, 27.35, 21.82, 21.70, 12.71.
results showed retainability of optical rotation direction by the
Preparation of (S)-(+)-methoxetamine hydrochloride acid free base and the HCl salt form. This suggested the presence of a
salt. (S)-(+)-Methoxetamine (+)-O,O0-di-p-toluoyl-D-tartarate common most stable conformer for both of the free base and
(1.76 g, 2.77 mmol) was treated with NaHCO3 aqueous solution the HCl salt forms. Conformational analysis confirmed this
and extracted three times with toluene, dried (anhydrous suggestion indicating that both the free base and salt forms
MgSO4) and evaporated under reduced pressure. The residue share axial-methoxyphenyl conformation of the most abundant
was treated with hydrochloric acid (1.5 eq. of 1.25 M methanolic stable conformers. Comparing experimental to calculated ECD
solution) and stirred at 0 1C for 30 minutes. After evaporation spectra unambiguously assigned the absolute configuration of
under reduced pressure, it was re-dissolved in methanol (4 mL) both free base and HCl forms of the resolved MXE enantiomers.
and recrystallized from diethyl ether at 0 1C to afford (S)-(+)-
methoxetamine hydrochloride acid salt (0.72 g, 91% yield, 99%
Statement of contributions
ee, Chiralpak IF-3, EtOH/MeOH/DEA = 50/50/0.1, 0.5 mL minꢀ1
rt: 5.30 min).
,
Y. S. Lee and J. H. Cheong conceived and designed the study;
A. H. E. Hassan, K. W. Lee, Y. Jeong, S. Yoon, C. J. Lee and H. R.
Jeon performed the experiments; K. W. Lee and Y. Jeong
contributed to acquisition of analytical data; S.-H. Kim, S. W.
Chang, J.-Y. Kim and D. S. Jang analyzed data; Y. S. Lee, A. H. E.
Hassan and J. H. Cheong drafted and revised the article.
White solid; mp: 235–245 1C; [a]2D5: +14.71 (c 1.0, EtOH,
hydrochloride salt); [a]2D5: +1651 (c 0.1, H2O, hydrochloride salt);
1
[a]2D5: +11.81 (c 0.5, EtOH, free base); H NMR (500 MHz, D2O)
d 7.51 (t, J = 8.0 Hz, 1H), 7.15 (dd, J = 8.6, 1.7 Hz, 1H), 7.01 (d, J =
8.0 Hz, 1H), 6.95 (s, 1H), 3.84 (s, 3H), 3.19 (d, J = 13.7 Hz, 1H),
2.87 (dt, J = 19.5, 7.2 Hz, 1H), 2.57–2.48 (m, 3H), 2.04–1.92 (m,
3H), 1.80–1.67 (m, 2H), 1.14 (t, J = 7.4 Hz, 3H); 13C NMR
(125 MHz, D2O) d 209.01, 160.08, 131.35, 131.06, 120.80, 115.92,
114.12, 72.03, 55.56, 39.01, 37.13, 32.37, 27.38, 21.18, 10.65.
Conformational analysis and calculation of electronic circular
dichroism (ECD) spectra. The 3D models of MXE free base and
HCl forms were built using the Chem 3D program. A conforma-
tional search was performed using the systematic (ring + spin)
stepped method as implemented in Spartan’14 software (Wave-
function, Inc., Irvin, CA, USA; 2014) using the Merk molecular
force field (MMFF). The algorithm parameters were set to
exclude conformers with energy outside a window of 40 kJ molꢀ1
above the energy of the global minimum conformation.
Geometry optimization of the obtained conformers was
subsequently performed at the B3LYP/6-31+G(d,p) level by the
Gaussian 09 software (Revision E.01; Gaussian, Inc., Wallingford,
CT, USA; 2009). Time-dependent density functional theory
(TDDFT) electronic circular dichroism (ECD) calculations of the
optimized conformers were carried out at the CAM-B3LYP/SVP
level with a conductor-like polarizable continuum solvent model
(CPCM) where MeCN was the solvent model (Gaussian 09
software).
Conflicts of interest
There are no conflicts to declare.
Acknowledgements
We are thankful to Ms Ju-Hyeon Jin from CTK (Chiral Technol-
ogy Korea) for her help in the chiral HPLC analysis of com-
pounds. This research was supported by the Ministry of Food
and Drug Safety (#20201063).
References
˜
1 C. J. Pena and E. J. Nestler, Prog. Mol. Biol. Transl. Sci., 2018,
157, 41–66.
2 M. Shadrina, E. A. Bondarenko and P. A. Slominsky, Front.
Psychiatry, 2018, 9, 334.
3 G. S. Malhi and J. J. Mann, Lancet, 2018, 392, 2299–2312.
4 J. Gordon, New Hope for Treatment-Resistant Depression:
Guessing Right on Ketamine, https://www.nimh.nih.gov/
about/director/messages/2019/new-hope-for-treatment-
resistant-depression-guessing-right-on-ketamine.shtml,
(accessed January 2021).
5 B. Kadriu, L. Musazzi, I. D. Henter, M. Graves, M. Popoli
and C. A. Zarate Jr, Int. J. Neuropsychopharmacol., 2019, 22,
119–135.
Conclusions
This work addressed the development of a chiral resolution
protocol capable of providing gram scale quantities of both
(R)- and (S)-MXE. First, racemic MXE was synthesized efficiently
following a modified scheme affording 5.5-fold the overall yield
of the reported synthetic method (24.2% relative to the 4.4%
6 J. H. Krystal, G. Sanacora, H. Blumberg, A. Anand,
D. S. Charney, G. Marek, C. N. Epperson, A. Goddard and
G. F. Mason, Mol. Psychiatry, 2002, 7(Suppl. 1), S71–S80.
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem., 2021, 45, 4354ꢀ4364 | 4363

View Article Online
NJC
Paper
7 A. Adell, Biomolecules, 2020, 10, 947.
22 B. L. Roth, S. Gibbons, W. Arunotayanun, X.-P. Huang,
V. Setola, R. Treble and L. Iversen, PLoS One, 2013,
8, e59334.
8 A. Haj-Mirzaian, N. Kordjazy, A. Haj-Mirzaian, S. Ostadhadi,
M. Ghasemi, S. Amiri, M. Faizi and A. Dehpour, Psychophar-
macology, 2015, 232, 3551–3561.
˜
23 C. J. Botanas, J. Bryan de la Pena, R. J. Custodio, I. Joy Dela
˜
9 M. K. Seo, L. T. Hien, M. K. Park, A. J. Choi, D.-H. Seog,
S.-H. Kim, S. W. Park and J. G. Lee, Sci. Rep., 2020, 10, 993.
10 C. Yang, W. Zhou, X. Li and J. Yang, Expert Opin. Invest.
Drugs, 2012, 21, 583–584; author reply 584–585.
11 D. Bleakman, A. Alt and J. M. Witkin, CNS Neurol. Disord.:
Drug Targets, 2007, 6, 117–126.
12 C. F. Zorumski, P. Nagele, S. Mennerick and C. R. Conway,
Front. Psychiatry, 2015, 6, 172.
13 L. R. Aleksandrova, A. G. Phillips and Y. T. Wang,
J. Psychiatry Neurosci., 2017, 42, 222–229.
Pena, M. Kim, T. Woo, H. J. Kim, H. I. Kim, M. Chang Cho,
Y. S. Lee and J. H. Cheong, Neuropharmacology, 2017, 126,
121–127.
24 B. Jurasek, M. Himl, R. Jurok, K. Hajkova, A. Vobinuskova,
P. Rezanka and M. Kuchar, RSC Adv., 2017, 7, 56691–56696.
25 J. Jose, S. A. Gamage, M. G. Harvey, L. J. Voss, J. W. Sleigh
and W. A. Denny, Bioorg. Med. Chem., 2013, 21, 5098–5106.
26 A. Moghimi, S. Rahmani, R. Zare and M. Sadeghzadeh,
Synth. Commun., 2014, 44, 2021–2028.
27 Y. A. Tawar, R. M. Patil, R. S. Lokhande and D. H. Ner,
Paripex Indian J. Res., 2017, 6, 201–204.
14 N. Llamosas, L. Perez-Caballero, E. Berrocoso, C. Bruzos-
Cidon, L. Ugedo and M. Torrecilla, Prog. Neuropsychophar- 28 B. Chankvetadze, N. Burjanadze, J. Breitkreutz,
¨
macol. Biol. Psychiatry, 2019, 88, 243–252.
15 J. McConathy and M. J. Owens, Prim. Care Companion J. Clin.
Psychiatry, 2003, 5, 70–73.
K. Bergander, D. Bergenthal, O. Kataeva, R. Frohlich,
H. Luftmann and G. Blaschke, J. Sep. Sci., 2002, 25,
1155–1166.
16 M. Eichelbaum, Arch. Toxicol., Suppl., 1995, 17, 514–521.
17 L. A. Nguyen, H. He and C. Pham-Huy, Int. J. Biomed. Sci.,
2006, 2, 85–100.
18 M. W. Tyler, H. B. Yourish, D. F. Ionescu and S. J. Haggarty,
ACS Chem. Neurosci., 2017, 8, 1122–1134.
19 R. Kohrs and M. E. Durieux, Anesth. Analg., 1998, 87,
1186–1193.
20 M. Narita, K. Yoshizawa, K. Aoki, M. Takagi, M. Miyatake
and T. Suzuki, Addict. Biol., 2001, 6, 373–376.
21 P. Zanos, R. Moaddel, P. J. Morris, L. M. Riggs,
29 P. Hakey, W. Ouellette, J. Zubieta and T. Korter, Acta
Crystallogr., Sect. E: Struct. Rep. Online, 2008, 64, o1487.
30 A. T. Fischer, R. N. Compton and R. M. Pagni, J. Phys. Chem.
A, 2006, 110, 7067–7071.
31 C. Yang, Y. Shirayama, J. C. Zhang, Q. Ren, W. Yao, M. Ma,
C. Dong and K. Hashimoto, Transl. Psychiatry, 2015, 5, e632.
32 R. Lafont, C. Blais, J. Harmatha and I. D. Wilson, in
Encyclopedia of Separation Science, ed. I. D. Wilson, Aca-
demic Press, Oxford, 2000, pp. 2631–2643, DOI: 10.1016/B0-
12-226770-2/06431-0.
J. N. Highland, P. Georgiou, E. F. R. Pereira, 33 J. Autschbach, L. Nitsch-Velasquez and M. Rudolph, Top.
E. X. Albuquerque, C. J. Thomas, C. A. Zarate, Jr. and
Curr. Chem., 2011, 298, 1–98.
T. D. Gould, Pharmacol. Rev., 2018, 70, 621–660.
34 G. Pescitelli and T. Bruhn, Chirality, 2016, 28, 466–474.
4364 | New J. Chem., 2021, 45, 4354ꢀ4364
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021