Positive Modulators of the N-Methyl- d -aspartate Receptor: Structure-Activity Relationship Study of Steroidal 3-Hemiesters

Article abstract of DOI:10.1021/acs.jmedchem.8b00255

Here, we report the synthesis of pregn-5-ene and androst-5-ene dicarboxylic acid esters and explore the structure-activity relationship (SAR) for their modulation of N-methyl-d-aspartate receptors (NMDARs). All compounds were positive modulators of recombinant GluN1/GluN2B receptors (EC₅₀ varying from 1.8 to 151.4 μM and E_max varying from 48% to 452%). Moreover, 10 compounds were found to be more potent GluN1/GluN2B receptor modulators than endogenous pregnenolone sulfate (EC₅₀ = 21.7 μM). The SAR study revealed a relationship between the length of the residues at carbon C-3 of the steroid molecule and the positive modulatory effect at GluN1/GluN2B receptors for various D-ring modifications. A selected compound, 20-oxo-pregnenolone hemiadipate, potentiated native NMDARs to a similar extent as GluN1/GluN2A-D receptors and inhibited AMPARs and GABA_AR responses. These results provide a unique opportunity for the development of new steroid based drugs with potential use in the treatment of neuropsychiatric disorders involving hypofunction of NMDARs.

Full text of DOI:10.1021/acs.jmedchem.8b00255

Subscriber access provided by Kaohsiung Medical University
Article
Positive Modulators of the N-Methyl-D-Aspartate Receptor:
Structure-Activity Relationship Study on Steroidal 3-Hemiesters
Barbora Krausova, Barbora Slavikova, Michaela Nekardova, Pavla Hubalkova,
Vojtech Vyklicky, Hana Chodounska, Ladislav Vyklicky, and Eva Kudova
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b00255 • Publication Date (Web): 30 Apr 2018
Downloaded from http://pubs.acs.org on May 1, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 48
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Positive Modulators of the NꢀMethylꢀDꢀAspartate
Receptor: StructureꢀActivity Relationship Study on
Steroidal 3ꢀHemiesters
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
1,3
1,4
Barbora Krausova , Barbora Slavikova , Michaela Nekardova , Pavla Hubalkova , Vojtech
1
2
1
2,*
Vyklicky , Hana Chodounska , Ladislav Vyklicky , Eva Kudova
1
Institute of Physiology of the Czech Academy of Sciences, Videnska 1083, Prague 4, 142 20,
Czech Republic
2
Institute of Organic Chemistry and Biochemistry of the Czech Academy of Sciences,
Flemingovo nam. 2, Prague 6 – Dejvice, 166 10, Czech Republic
3
Faculty of Mathematics and Physics, Charles University in Prague, Ke Karlovu 3, Prague 2,
121 16, Czech Republic
4
Third Faculty of Medicine, Charles University in Prague, Ruska 87, Prague 10, 100 00, Czech
Republic
KEYWORDS. Neurosteroid, pregnꢀ5ꢀene, androstꢀ5ꢀene, ester, structureꢀactivity
relationship, NMDA receptor
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 48
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ABSTRACT. Here, we report the synthesis of pregnꢀ5ꢀene and androstꢀ5ꢀene dicarboxylic
acid esters and explore the structureꢀactivity relationship (SAR) for their modulation of
NꢀmethylꢀDꢀaspartate receptors (NMDARs). All compounds were positive modulators of
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
recombinant GluN1/GluN2B receptors (EC varying from 1.8 to 151.4 µM and E varying
max
5
0
from 48 to 452 %). Moreover, 10 compounds were found to be more potent GluN1/GluN2B
receptor modulators than endogenous pregnenolone sulfate (EC = 21.7 µM). The SAR study
50
revealed a relationship between the length of the residues at carbon Cꢀ3 of the steroid molecule
and the positive modulatory effect at GluN1/GluN2B receptors for various Dꢀring modifications.
A selected compound – 20ꢀoxoꢀpregnenolone hemiadipate – potentiated native NMDARs to a
similar extent as GluN1/GluN2AꢀD receptors, and inhibited AMPARs and GABA R responses.
A
These results provide a unique opportunity for the development of new steroidꢀbased drugs with
potential use in the treatment of neuropsychiatric disorders involving hypofunction of NMDARs.
ACS Paragon Plus Environment

Page 3 of 48
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Introduction.
NꢀmethylꢀDꢀaspartate receptors (NMDARs) are glutamateꢀgated, calciumꢀpermeable ion
channels that are activated during excitatory synaptic transmission and are implicated in various
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
forms of synaptic plasticity, which underlies learning and memory processes. Nextꢀgeneration
genome sequencing allowed identification of mutations in GRIN genes encoding for human
3
NMDAR subunits that have been associated with various neurodevelopmental disorders. The
4
analysis of mutated receptors revealed various forms of trafficking and functional defects.
Positive allosteric modulators (PAMs) that increase the activity of NMDARs may provide a
therapeutic aid for patients suffering from neuropsychiatric disorders where NMDARs
5
6
hypofunction is thought to be involved, such as intellectual disability, autism spectrum disorder
7
or schizophrenia.
8
Neurosteroids are endogenous steroidal compounds that modulate the activity of NMDARs.
Despite structural similarity between different neurosteroids, they can exhibit opposite effects at
NMDARs, apparently mediated by different molecular mechanisms. The inhibitory effect of
neurosteroids on NMDARs is dependent upon a 3αꢀnegatively charged substituent and 5βꢀ
9
stereochemistry. This class of neurosteroids is represented by endogenous pregnanolone sulfate
1
0
(20ꢀoxoꢀ5βꢀpregnanꢀ3αꢀyl sulfate, Figure 1A). In contrast, a 3βꢀnegatively charged moiety in
5
combination with ∆ ꢀ stereochemistry (a double bond between Cꢀ5 and Cꢀ6) favors potentiation
9
a,11
of NMDARs.
This class of neurosteroids is represented by endogenous pregnenolone sulfate
1
2
(20ꢀoxoꢀpregnꢀ5ꢀenꢀ3βꢀyl sulfate, compound 1, PES, Figure 1B). PES is an abundantly
occurring neurosteroid synthesized de novo in the central nervous system, which exhibits
different modulatory effects on several types of receptors: specifically potentiating the responses
1
2
elicited by NMDARs while inhibiting currents mediated by γꢀaminobutyric acid type A
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 48
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
13
14
receptors
(GABA Rs),
glycine
receptors
and
αꢀaminoꢀ3ꢀhydroxyꢀ5ꢀmethylꢀ4ꢀ
A
1
2
isoxazolepropionic acid receptors (AMPARs). A memory enhancement effect after the
1
5
administration of PES in vivo has been reported.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1. Structure of (A) pregnanolone sulfate, (B) pregnenolone sulfate (1, PES) and (C)
hemiester analogues of PES.
A structurally distinct class of compounds that act as PAMs of NMDARs, represented by
(
3ꢀchlorophenyl)(6,7ꢀdimethoxyꢀ1ꢀ((4ꢀmethoxyphenoxy)methyl)ꢀ3,4ꢀdihydroisoquinolinꢀ2(1H)ꢀ
yl)methanone (CIQ), was identified by an extensive study involving 100,000 substances focused
1
6
on selective action at NMDARs containing GluN2C or GluN2D subunits. The most effective
CIQ analogs have EC values in the submicromolar range and induce an approximately twoꢀfold
5
0
17
potentiation of GluN1/GluN2D receptor responses. By replacing the dimethyl groups on the
1
8
Cꢀring with one isopropoxyl group, CIQ analogs can be targeted to the GluN2B subunit.
However, the analogue that acts solely at GluN2B subunits exhibits only moderate efficacy and
18
potency (Emax = 152%; EC = 7.2 ꢁM).
50
The SAR for steroidꢀbased PAMs of NMDARs is only poorly understood and therefore needs
to by further characterized, especially in the context of the development of new drug candidates
for the prevention and/or treatment of neuropsychiatric disorders and ageꢀrelated memory
ACS Paragon Plus Environment

Page 5 of 48
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
4,19
deficits. Here, we describe a SAR study on hemiester analogues 2ꢀ24 (Figure 1C and Table
1
) with varying lengths of Cꢀ3 substituents and different steroidal Dꢀring modifications,
primarily focusing on fineꢀtuning their modulatory effect on NMDARs. The biological activity
of compounds 2ꢀ24 was evaluated on recombinant GluN1/GluN2B receptors expressed in human
embryonic kidney (HEK293) cells. The effect of a selected compound, 20ꢀoxoꢀpregnꢀ5ꢀene
hemiadipate (6), was further analyzed on recombinant GluN1/GluN2AꢀD receptors, and native
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
NMDARs, AMPARs, and GABA Rs expressed in hippocampal neurons.
A
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 48
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Results and Discussion
Table 1. Structures of PES analogues
̶ compounds 2ꢀ24.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Chemistry.
The synthesis of compounds 1ꢀ13 is summarized in Scheme 1. Compounds 1ꢀ8 were prepared
from commercially available 20ꢀoxoꢀpregnꢀ5ꢀenꢀ3βꢀol (pregnenolone, compound 25, Steraloids,
Newport, RI, USA). Compounds 9ꢀ13 were prepared from 3βꢀhydroxyꢀpregnꢀ5ꢀene (26), which
2
0
was obtained by the Zn/TMSClꢀmediated Clemmensen reduction of pregnenolone 25 (78%
yield). Pyridinium salt of pregnenolone sulfate (1) was prepared by the treatment of
pregnenolone (25) with a sulfur trioxideꢀpyridine complex in CHCl (89% yield). Note that
3
compound 1 is not commercially available as is sodium salt and in our experiments exerts better
solubility and chemical stability. Compound 2 was prepared by the treatment of pregnenolone 25
ACS Paragon Plus Environment

Page 7 of 48
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
with oxalyl chloride in DCM, Et N and DMF (36% yield). Compounds 3 and 9 were prepared by
3
the treatment of 25 and 26, respectively, with Meldrum´s acid in toluene at 80 °C (37% and 32%
yield, respectively). Compounds 4, 10 and 5, 11 were prepared by the treatment of 25 and 26
respectively with succinic or glutaric anhydride and DMAP in pyridine at 110 °C, yielding
hemiesters 4 (52% yield), 5 (51% yield), 10 (43% yield), and 11 (44% yield). Compounds 6ꢀ8
and 12, 13 were prepared by the treatment of 25 and 26, respectively with adipic, pimelic, or
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
suberic
acid,
1ꢀethylꢀ3ꢀ(3ꢀdimethylaminopropyl)carbodiimide
(EDCI),
N,Nꢀ
diisopropylethylamine (DIPEA) and DMAP in DCM, yielding hemiesters 6 (42% yield), 7 (47%
yield), 8 (44% yield), 12 (21% yield), and 13 (35% yield).
a
Scheme 1. Synthesis of pregnꢀ5ꢀene derivatives (1ꢀ13)
a
Reagents and conditions: (i) sulfur trioxideꢀpyridine complex, CHCl , pyridine; (ii) Zn,
3
TMSCl, MeOH, DCM; (iii) for compound 2: oxalyl chloride, Et N, DCM, DMF; for compound
3
3
and 9: Meldrum´s acid, toluene, 80°C; for compound 4 and 10: succinic anhydride, DMAP,
pyridine, 110 °C; for compound 5 and 11: glutaric anhydride, DMAP, pyridine, 110 °C; for
compounds 6 and 12: adipic acid, EDCI, DIPEA, DMAP, DCM; for compound 7 and 13:
pimelic acid, EDCI, DIPEA, DMAP, DCM; for compound 8: suberic acid, EDCI, DIPEA,
DMAP, DCM.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 48
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The synthesis of compounds 14ꢀ24 is summarized in Scheme 2. Compounds 19ꢀ24 were
prepared from commercially available 17ꢀoxoꢀandrostꢀ5ꢀenꢀ3βꢀol (27, Steraloids, Newport, RI,
2
1
USA). Compounds 14ꢀ18 were prepared from 3βꢀhydroxyꢀandrostꢀ5ꢀene (28), which was
2
0
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
obtained by the Zn/TMSClꢀmediated Clemmensen reduction of compound 27 (75% yield).
Compound 14 was prepared by the treatment of 28 with Meldrum´s acid in toluene at 80 °C
(
13% yield). Compounds 15, 16 and 19 were prepared by the treatment of 27 and 28 respectively
with succinic or glutaric anhydride and DMAP in pyridine at 110 °C, yielding hemiesters 15
67% yield), 16 (42% yield), and 19 (58% yield). Compounds 17, 18 and 20ꢀ24 were prepared by
(
the treatment of 27 and 28, respectively with adipic, pimelic, suberic, azelaic, or sebacic acid,
EDCI, DIPEA and DMAP in DCM, yielding hemiesters 17 (35% yield), 18 (42% yield), 20
(31% yield), 21 (35% yield), 22 (50% yield), 23 (24% yield), and 24 (27% yield). Although
these unoptimized reactions provided low yields of the desired products, we have demonstrated
the feasibility of this approach to obtain compounds 14ꢀ24.
a
Scheme 2. Synthesis of androstꢀ5ꢀene derivatives (14ꢀ24)
a
Reagents and conditions: (i) Zn, TMSCl, MeOH, DCM; (ii) for compound 14: Meldrum´s
acid, toluene, 80 °C; for compound 15: succinic anhydride, DMAP, pyridine, 110 °C; for
compound 16 and 19: glutaric anhydride, DMAP, pyridine, 110 °C; for compound 17 and 20:
adipic acid, EDCI, DIPEA, DMAP, DCM; for compound 18 and 21: pimelic acid, EDCI,
ACS Paragon Plus Environment

Page 9 of 48
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
DIPEA, DMAP, DCM; for compound 22: suberic acid, EDCI, DIPEA, DMAP, DCM; for
compound 23: azelaic acid, EDCI, DIPEA, DMAP, DCM; for compound 24: sebacic acid,
EDCI, DIPEA, DMAP, DCM.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Biological Activity.
To investigate the effect of PES (1) and its analogues (compounds 2ꢀ24; Figure 1C and
Table 1) on the activity of NMDARs, electrophysiological measurements were performed on
HEK293 cells that were coꢀtransfected with cDNAs containing genes encoding for the rat
GluN1ꢀ1a and GluN2B subunits. The degree of modulation (E; potentiation/inhibition) for the
newly synthesized compounds was determined using the following formula:
(
ꢁₑꢂꢁₐ)
ꢀ =
× 100
Equation (1),
ꢁₐ
where I is the value of the current amplitude during glutamate and compound coꢀapplication
e
and I is the current amplitude value for glutamate application. The degree of potentiation (%)
a
was determined for five compound doses, differing 100ꢀfold in their concentration range, in
individual cells, and data were fit to the following equation:
ꢗ
ꢖ ꢘꢃ
ꢉꢊꢋꢌ
ꢃꢃꢃꢃꢃꢀ =ꢃEꢄꢅꢆꢙꢇ1 + ꢈ
Equation (2),
ꢍꢎꢏꢐꢑꢏꢒꢓꢔꢕ
where Emax is the maximal value of potentiation induced by a saturating concentration of the
compound, EC is the concentration of the compound that produces halfꢀmaximal potentiation
50
of the agonistꢀevoked current, [compound] is the compound concentration and h is the apparent
Hill coefficient.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 48
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The results are presented as mean ± SEM, with n equal to the number of studied cells.
Unpaired Student’s tꢀtest was used to perform a statistical comparison between two treatment
groups and Oneꢀway Analysis of Variance (ANOVA) for multiple comparisons (unless
otherwise stated, a value of P ≤ 0.05 was used for the determination of significance).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Compounds 2ꢀ24 Potentiate Recombinant GluN1/GluN2B Receptor Responses.
We have evaluated a relationship between the steroid structure and its modulatory effect on
GluN1/GluN2B receptors for compounds 2ꢀ24. Figure 2 shows that compound 6 (30 μM)
potentiates GluN1/GluN2B receptor responses. The degree of potentiation was affected by the
relative timing of compound and glutamate application and dependent on glutamate
concentration. The responses induced by 1 µM glutamate were potentiated by 136 ± 8% (n = 10)
when compound 6 was coꢀapplied with the agonist. The potentiation was only 75 ± 1% (n = 6)
when compound 6 was coꢀapplied with a saturating concentration of glutamate (1 mM). The
responses to 1 mM glutamate were potentiated by 134 ± 7% (n = 8) when compound 6 was preꢀ
applied prior to glutamate (not significantly different from the potentiation induced at a low
concentration of glutamate; unpaired Student’s tꢀtest, P = 0.398). These results are in agreement
2
2
with the disuseꢀdependent effect of PES at NMDARs. In subsequent experiments we used
compound coꢀaplication with 1 µM glutamate.
ACS Paragon Plus Environment

Page 11 of 48
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 2. The potentiating effect of compound 6 at GluN1/GluN2B receptors. Examples of
traces obtained from HEK293 cells transiently expressing GluN1/GluN2B receptors. (A)
Compound 6 (30 ꢁM) was applied simultaneously with 1 ꢁM glutamate (the duration of
compound 6 and glutamate application is indicated by filled and open bars, respectively). (B)
Compound 6 (30 ꢁM) was either coꢀapplied with 1 mM glutamate (on the left), or preꢀapplied
before the receptor was activated by 1 mM glutamate (control and potentiated response on the
right). (D) Bar graph shows the degree of potentiation by compound 6 (30 ꢁM) when coꢀapplied
with 1 ꢁM glutamate (filled bar), coꢀapplied with 1 mM glutamate (open bar) and preꢀapplied
prior to 1 mM glutamate (open bar). Asterisks mark significance at the level of P ≤ 0.001.
In initial experiments, we analyzed the effect of compounds 2ꢀ24 at a concentration of 30 µM
on GluN1/GluN2B receptors. Figure 3 shows the SAR for the series of 20ꢀoxoꢀpregnꢀ5ꢀenes
(2ꢀ8), pregnꢀ5ꢀenes (9ꢀ13), androstꢀ5ꢀenes (14ꢀ18), and 17ꢀoxoꢀandrostꢀ5ꢀenes (19ꢀ24). The
analysis of the potentiating effect indicates that there is an optimal length of the substituents at
Cꢀ3 in combination with structural modification at Cꢀ17 of the steroidal Dꢀring for acquiring the
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 48
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
desired biological effect. Once the 20ꢀoxo substituent of 20ꢀoxoꢀPEꢀhAdi (6) was removed, the
most efficacious PAM was PEꢀhGlu (11), which has a Cꢀ3 moiety with five carbons.
Analogously, when the steroidal skeleton does not bear the Cꢀ17 acyl, the most efficacious PAM
was ANDꢀhSuc (15), which has a Cꢀ3 moiety with four carbons. Such common structural
features for compounds 2ꢀ24 may characterize the pharmacophore of steroidal structural
requirements for the optimal potentiating effect at NMDARs. Previously, compounds 2, 4, and 5
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
+
9a
were found to enhance NMDARsꢀinduced Ca accumulation in cultured hippocampal neurons.
ACS Paragon Plus Environment

Page 13 of 48
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 3. The effect of compounds 2ꢀ24 on current responses of GluN1/GluN2B receptors. Bar
graphs show changes in response to the application of compounds 2ꢀ24 at a concentration of
3
0 ꢁM. The numbers above the bars indicate the number of carbons of the particular hemiester
moiety. (A) 20ꢀOxoꢀpregnꢀ5ꢀenꢀ3βꢀyl hemiesters. (B) Pregnꢀ5ꢀenꢀ3βꢀyl hemiesters. (C) Androstꢀ
ꢀenꢀ3βꢀyl hemiesters. (D) 17ꢀOxoꢀandrostꢀ5ꢀenꢀ3βꢀyl hemiesters. Differences in mean values
within the structurally related groups (2ꢀ8, 9ꢀ13, 14ꢀ18 and 19ꢀ24) were statistically significant
One Way ANOVA; P < 0.001).
5
(
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 48
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Next, we have performed doseꢀresponse analysis for the effect of compounds 1ꢀ24 on
GluN1/GluN2B receptor responses induced by 1 µM glutamate (Figure 4). Fit to the
Equation (2) provided information on the compounds 1ꢀ24 efficacy and potency. The Emax and
EC50 values for compounds 1ꢀ24 are shown in Table 2. Compounds 1ꢀ8 and 21ꢀ24 were
measured up to a concentration of 100 µM and compounds 19 and 20 up to a concentration of
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
00 µM. Compounds 9ꢀ18 were measured up to a concentration of 30 µM due to their limited
solubility. Our data show that 17 out of 23 newly synthesized compounds had increased efficacy
and/or potency at GluN1/GluN2B receptors as compared to endogenous PES (1).
ACS Paragon Plus Environment

Page 15 of 48
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 4. The effect of compounds 1, 6, 11, 15 and 23 on GluN1/GluN2B receptors. Graphs
show the doseꢀresponse curves for the effect of the following compounds: (A) PES (1; in red)
and 20ꢀOxoꢀPEꢀhAdi (6), (B) PEꢀhGlu (11), (C) ANDꢀhSuc (15) and (D) 17ꢀOxoꢀANDꢀhAze
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(23) at GluN1/GluN2B receptors. The red line in (BꢀD) repeats the fitted function for PES (1),
for comparison. Data points are averaged potentiation values from at least four independent
measurements. Error bars represent SEM. The degree of potentiation of glutamateꢀinduced
responses recorded in the presence of compound 1 (3ꢀ100 ꢁM), 6 (1ꢀ30 ꢁM), 11 (0.3ꢀ30 ꢁM), 15
(0.3ꢀ30 ꢁM) and 23 (1ꢀ100 ꢁM) was determined in individual cells and data were fitted to the
Equation (2). Insets show examples of traces obtained from HEK293 cells coꢀtransfected with
cDNAs encoding for GluN1 and GluN2B subunits. Compounds 1, 6, 11, 15 and 23 were applied
simultaneously with 1 ꢁM glutamate (duration of compound and glutamate application is
indicated by filled and open bars, respectively, PES (1) in red).
Table 2. Summary of the doseꢀresponse analysis data for the potentiating effect of compounds
1ꢀ24 at GluN1/GluN2B receptor responses.
Emax ± SEM
%)
EC ± SEM
50
Steroid
h ± SEM
n
(
(µM)
PES (1)
116 ± 10
21.7 ± 1.6
1.5 ± 0.1
1.3 ± 0.0
1.3 ± 0.1
1.5 ± 0.2
1.4 ± 0.1
8
5
5
4
4
2
0ꢀOxoꢀPEꢀhOxa (2)
93 ± 25
48.6 ± 12.4*
39.8 ± 10.3
27.6 ± 12.1
19.0 ± 3.4
2
0ꢀOxoꢀPEꢀhMal (3)
0ꢀOxoꢀPEꢀhSuc (4)
20ꢀOxoꢀPEꢀhGlu (5)
0ꢀOxoꢀPEꢀhAdi (6)
0ꢀOxoꢀPEꢀhPim (7)
108 ± 15
177 ± 46*
183 ± 25*
151 ± 16
191 ± 63
2
2
8.5 ± 1.0**
18.0 ± 7.9
1.6 ± 0.1 10
1.4 ± 0.2
2
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 48
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
0ꢀOxoꢀPEꢀhSub (8)
156 ± 30
P = 0.346
19.3 ± 1.7
1.5 ± 0.1
P = 0.494
1.5 ± 0.2
5
Oneꢀway ANOVA
P < 0.001
PEꢀhMal (9)
PEꢀhSuc (10)
71 ± 7*
133 ± 19
10.7 ± 0.8**
9.9 ± 5.2*
7
6
5
4
4
1.4 ± 0.2
1.8 ± 0.3
1.5 ± 0.4
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
PEꢀhGlu (11)
209 ± 22*
60 ± 12*
9.9 ± 3.7*
PEꢀhAdi (12)
17.2 ± 4.5
1.4 ± 0.4
P = 0.820
PEꢀhPim (13)
48 ± 11*
26.2 ± 8.1
Oneꢀway ANOVA
ANDꢀhMal (14)
ANDꢀhSuc (15)
ANDꢀhGlu (16)
ANDꢀhAdi (17)
ANDꢀhPim (18)
Oneꢀway ANOVA
7ꢀOxoꢀANDꢀhGlu (19)
P = 0.002
236 ± 27**
452 ± 46*
348 ± 32**
191 ± 10**
109 ± 13
P = 0.080
1
.4 ± 0.2
.5 ± 0.1
4.7 ± 1.0**
7.4 ± 0.4**
5.2 ± 1.2**
2.8 ± 0.5**
1.8 ± 0.3**
P < 0.001
4
5
7
5
4
1
1.7 ± 0.2
1
.7 ± 0.5
.8 ± 0.3
1
P = 0.892
1.2 ± 0.1
1.4 ± 0.1
P <0.001
154 ± 34
1
151.4 ± 32.7**
108.9 ± 20.2**
82.4 ± 16.9**
49.1 ± 13.7**
38.2 ± 14.5
16.1 ± 0.9*
P < 0.001
4
7
7
5
5
5
1
7ꢀOxoꢀANDꢀhAdi (20)
180 ± 17*
187 ± 27*
189 ± 35*
226 ± 45*
165 ± 11*
P = 0.654
1.2 ± 0.1
1.4 ± 0.2
1.3 ± 0.2
17ꢀOxoꢀANDꢀhPim (21)
1
7ꢀOxoꢀANDꢀhSub (22)
7ꢀOxoꢀANDꢀhAze (23)
1
1.6 ± 0.2
P = 0.661
1
7ꢀOxoꢀANDꢀhSeb (24)
Oneꢀway ANOVA
The degree of potentiation of glutamateꢀinduced responses, recorded in the presence of a given
compound at concentrations of 0.3ꢀ30 ꢁM (for compounds 9ꢀ18), 1ꢀ100 ꢁM (for compounds 1ꢀ8
and 21ꢀ24), and 1ꢀ300 ꢁM (for compounds 19 and 20) was determined in individual cells for at
least five compound doses and the data were fitted to the Equation (2). Statistical tests on the
potency of compounds (2ꢀ8, 9ꢀ13, 14ꢀ18 and 19ꢀ24) were performed on logEC and logHill
values (Oneꢀway ANOVA); P ≤ 0.05 was used for the determination of significance, as marked
in bold. Unpaired Student’s tꢀtest was used for statistical comparisons of Emax and logEC50
values for compounds 2ꢀ24 to PES (1;*P = 0.001− 0.050; **P < 0.001).
5
0
The series of 20ꢀoxoꢀpregnꢀ5ꢀenes (2ꢀ8) exhibits similar values of Emax within this structural
family (from 93% for the least efficacious compound 2 to 191% for the most efficacious
ACS Paragon Plus Environment

Page 17 of 48
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
compound 7). In contrast, the EC values for compounds 2ꢀ8 differ significantly, with the EC
50
5
0
of the most potent compound 6 (Figure 4A) being up to 6ꢀfold lower (8.5 µM) than that of the
least potent compound 2 (48.5 µM). The doseꢀresponse analysis of compounds 4ꢀ7 indicated that
at a concentration of 100 µM these compounds potentiated more than predicted from the fit of
their potentiating effect at concentrations of 1ꢀ30 ꢁM. The potentiating effect induced by these
compounds at a concentration of 100 µM was: 267 ± 47% (n = 4) for compound 4, 460 ± 17%
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(n = 4) for compound 5, 614 ± 56% (n = 10) for compound 6, and 458 ± 119% (n = 4) for
compound 7. The reason for this increased potentiation at high compound concentrations is
unknown and we did not analyze it in detail. Augmented potentiation was not observed for
compounds 1ꢀ3 and 8ꢀ24 (100 µM).
To evaluate the role of the 20ꢀoxo group at the steroid Dꢀring for biological activity at
NMDARs, a series of pregnꢀ5ꢀenes (9ꢀ13) was synthesized. Interestingly, within this structural
family, compounds exhibited more than 4ꢀfold difference in the Emax values (48% for the least
efficacious compound 13 compared to 209% for the most efficacious compound 11; Figure 4B),
with EC values varying from 9.9 µM for the most potent compound 11 to 26.2 µM for the least
5
0
potent compound 13.
Further, we analyzed a series of androstꢀ5ꢀenes (14ꢀ18) lacking a substituent at the position
Cꢀ17 of the steroidal Dꢀring. Compound 15 (ANDꢀhSuc) was established as the most efficacious
PAM within this structural family with Emax value of 452% (Figure 4C), approximately 4.5ꢀ
times higher than that of the least efficacious compound 18 (109%). In general, compounds
1
4ꢀ18 were recognized as the most potent PAMs of all the tested compounds 2ꢀ24, with EC50
values varying from 1.8 µM for the most potent compound 18 to 7.4 µM for the least potent
compound 15 from this series (Table 2). This structural modification led to a significant
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 48
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
improvement in potency for all compounds 14ꢀ18, suggesting that nonꢀsubstituted Dꢀring drives
the activity of PAM toward stronger potency.
Finally, we tested a series of 17ꢀoxo analogues (19ꢀ24) with 17ꢀoxo group at the steroidal
Dꢀring. 17ꢀOxo analogues exhibited similar values of Emax within this structural family (varying
from 154% for the least efficacious compound 19 to 226% for the most efficacious compound
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
3; Figure 4D). The potency of compounds 19ꢀ24 differs significantly, exhibiting up to 9ꢀfold
decrease in EC value for the most potent compound 24 (16.1 µM) compared to the least potent
5
0
compound 19 (151.4 µM).
Computational Estimate of Physicochemical Properties of Compounds 2ꢀ24.
Our recent results for steroids with inhibitory action at NMDARs showed the plasma
2
3
membrane as the route for the steroid to reach its binding site on the receptor. We could
speculate that steroids with a potentiating effect, which are structurally similar to the inhibitory
ones, may employ the same route. Therefore, we have analyzed the role of lipophilicity in the
potency of compounds at NMDARs.
Computational analysis was used to evaluate the lipophilic qualities of compounds 1ꢀ24. The
relevant physicochemical properties (
ꢀG , logP and logD values), which are commonly used
solv
2
4
for describing the compound lipophilicity,
were calculated by quantum mechanics
2
5
computational methods and with a physicochemical properties predictor. The computational
results are summarized in Table S1 (for details see Supporting Information). Compounds 5 and
1
4 show similar values of logD (1.29 and 1.28, respectively). However, the EC50 value for
compound 5 (19.0 µM) is 4ꢀtimes higher than for compound 14 (4.7 µM). Similarly, more
lipophilic compounds 11 and 22 (logD values of 2.47 and 2.42, respectively) show an
ACS Paragon Plus Environment

Page 19 of 48
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
approximately fiveꢀfold difference in EC50 values (Table 2). These results suggest that
compound lipophilicity and presumed solubility in the membrane play only a minor role in
determining the compound potency and indirectly indicate that potency is more dependent on the
specific (structural/binding) interactions between the compound and the NMDAR.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Effect of 20ꢀOxoꢀPEꢀhAdi (6) on Recombinant GluN1/GluN2AꢀD and Native Receptors.
Compound 6 was selected from the library of compounds 2ꢀ24 for further biological
evaluation. To investigate the subunit selectivity of compound 6, cDNAs coding for the GluN1
and GluN2AꢀD subunits were coꢀtransfected into HEK293 cells. Doseꢀresponse analysis for the
effect of compound 6 on GluN1/GluN2AꢀD receptors showed no significant difference in the
efficacy and potency at receptors differing in the subunit composition (Table 3).
Subsequently we compared the effect of compound 6 (30 ꢁM) and PES (100 μM) at
recombinant GluN1/GluN2AꢀD receptors. The results (Figure 5) indicate that compound 6 (30
μM) potentiated all tested subunit compositions of GluN1/GluN2AꢀD receptors (the degree of
potentiation was: 192 ± 21% for GluN1/GluN2A, 137 ± 11% for GluN1/GluN2B, 118 ± 17% for
GluN1/GluN2C, and 88 ± 11% for GluN1/GluN2D; n = 6). PES (100 μM) potentiated
GluN1/GluN2AꢀB receptor responses (107 ± 17% (n = 4) for GluN1/GluN2A and 78 ± 4%
(n = 13) for GluN1/GluN2B) while it inhibited GluN1/GluN2CꢀD receptor responses (ꢀ47 ± 4%
for GluN1/GluN2C and ꢀ48 ± 3% for GluN1/GluN2D; n = 4). This agrees well with the results of
previous experiments showing that PES has a dual potentiating and inhibitory action at
2
2,26
NMDARs.
The predominant effect of PES (potentiation of GluN1/GluN2AꢀB and inhibition
of GluN1/GluN2CꢀD) is the result of different efficacy rather than potency, which is higher for
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 48
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
26
GluN1/GluN2CꢀD receptors than for GluN1/GluN2AꢀB receptors. In our experiments we have
seen no evidence of a dual effect for compound 6.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 3. Summary of the doseꢀresponse analysis data for the potentiation effect of compound 6
on responses of GluN1/GluN2AꢀD receptors.
Emax ± SEM
(%)
EC ± SEM
50
Receptor
h ± SEM
n
(µM)
GluN1/GluN2A
GluN1/GluN2B
GluN1/GluN2C
GluN1/GluN2D
203 ± 43
157 ± 16
145 ± 39
115 ± 9
8.9 ± 1.4
1.7 ± 0.2
1.7 ± 0.1
1.7 ± 0.2
1.7 ± 0.3
7
8
6
5
9.1 ± 1.1
10.4 ± 3.5
13.0 ± 1.9
Oneꢀway
ANOVA
P = 0.363
P = 0.520
P = 0.997
The degree of potentiation of glutamateꢀinduced responses recorded in the presence of
compound 6 (1ꢀ30 ꢁM) was determined in individual cells, and data were fit to the Equation (2).
Statistical tests on the compound 6 potency were performed on logEC and logHill values (Oneꢀ
5
0
way ANOVA).
ACS Paragon Plus Environment

Page 21 of 48
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 5. Comparison of the effect of PES and compound 6 on GluN1/GluN2AꢀD receptor
responses. Examples of traces obtained from HEK293 cells transfected with cDNAs encoding
for GluN1 and GluN2AꢀD subunits. PES (100 ꢁM) and compound 6 (30 ꢁM) were applied
simultaneously with 1 ꢁM glutamate (the duration of the compound and glutamate application is
indicated by filled and open bars, respectively).
Native NMDARs expressed in cultured hippocampal neurons were potentiated (138 ± 11%,
n = 6; Figure 6) by compound 6 (30 ꢁM) to a similar extent as recombinant GluN1/GluN2B
receptors (136%; Figure 2A; P = 0.824). In contrast to the relatively pronounced potentiating
effect of compound 6 on native and recombinant NMDARs responses, compound 6 (30 µM) had
only a small inhibitory effect (ꢀ7 ± 2%, n = 6) on responses induced by selective AMPARs
agonist (AMPA 5 ꢁM) recorded in the presence of cyclothiazide (10 ꢁM) for reducing receptor
27
desensitization. Compound 6 (30 µM) inhibited responses induced by 5 ꢁM GABA (ꢀ90 ± 2%,
n = 5). These effects of compound 6 are qualitatively similar to PES effects at native NMDARs,
1
3,28
AMPARs, and GABA Rs.
A
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 48
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 6. The effect of compound 6 on native NMDARs, AMPARs, and GABA Rs. Examples
A
of traces obtained from cultured hippocampal neurons. Compound 6 (30 ꢁM) was applied
simultaneously with NMDA (10 ꢁM), AMPA (5 ꢁM) and GABA (5 ꢁM) (the duration of
compound and agonist application is indicated by filled and open bars, respectively).
Conclusions
5
In this study, we examined a library of compounds 2ꢀ24, bearing a Cꢀ3 hemiester moiety, ∆ ꢀ
double bond and various modifications at position Cꢀ17, in order to evaluate their ability to
modulate the activity of NMDARs. The results of our experiments indicate that the Cꢀ17
substituent of the Dꢀring can be structurally modified or fully degraded while maintaining a
positive modulatory effect of the steroid. We have also shown that the potentiating effect of these
compounds exhibits a dependency on the length of the Cꢀ3 substituent for each Dꢀring
modification. The most efficacious and potent modulators, respectively, from all tested
compounds (2ꢀ24), were compound 15, exhibiting the Emax value of 452%, and compound 18,
exhibiting the EC value of 1.8 µM. In addition, we have shown that the selected compound 6
50
ACS Paragon Plus Environment

Page 23 of 48
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
has a subunitꢀindependent effect at recombinant NMDARs, which is similar to that at native
NMDARs, and has only a minor inhibitory effect at AMPARs. Our data further indicate that
compound 6 is an inhibitor of native GABA Rs. In summary, we conclude that PES analogues
A
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
modified at the Cꢀ3 and/or at the Dꢀring offer new prospects for further optimization of
pharmacological and pharmacokinetic properties of these neuroactive compounds.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 48
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Experimetal Section
Chemistry.
General. Melting points were determined with a Hund/Wetzlar micromelting point apparatus
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(Germany) and are uncorrected. Optical rotations were measured in chloroform using an Autopol
˗
1
2
ꢀ1
IV (Rudolf Research Analytical, Flanders, USA). [α] values are given in deg (10 deg cm g ).
D
ꢀ
1
IR spectra were recorded in a Bruker IFS 55 spectrometer (wavenumbers in cm ). Proton and
carbon NMR spectra were measured in a Bruker AVANCEꢀ400 FT NMR spectrometer (400
MHz, 101 MHz) in CDCl , with tetramethylsilane as the internal standard. Chemical shifts are
3
given in ppm (δ scale). Coupling constants (J) and widths of multiplets (W) are given in Hz.
Highꢀresolution MS spectra were performed with a QꢀTof microspectrometer (Waters).
Elemental analysis was performed on a PerkinꢀElmer 2400 Series II CHNS/O Analyzer (USA).
Thinꢀlayer chromatography (TLC) was performed on silica gel (ICN Biochemicals). For column
chromatography, neutral silica gel 60 µm (Merck) was used. Analytical samples were dried over
phosphorus pentoxide at 50 °C/100 Pa. The purity of the final compounds was assessed by a
combination of NMR and on the basis of LCꢀHRꢀMS analysis or elemental analysis, and the
results showed they were greater than 95%.
Biological Activity.
Electrophysiological experiments were performed on HEK293 cells transfected with cDNA
containing GluN1ꢀ1a/GluN2AꢀD/GFP genes and on cultured hippocampal neurons as described
23a,29
previously.
Glutamateꢀinduced responses were recorded at a holding potential of ꢀ60 mV.
Wholeꢀcell voltageꢀclamp recordings were made with a patchꢀclamp amplifier (Axopatch 200B;
Axon Instruments. Inc., Foster City, CA) after series resistance (<10 Mꢁ) and capacitance
ACS Paragon Plus Environment

Page 25 of 48
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
compensation of 80ꢀ90%. For the application of test and control solutions, a microprocessorꢀ
controlled multibarrel fastꢀperfusion system was used, with a time constant of solution exchange
around the cells of ~10 ms. Agonistꢀinduced responses were lowꢀpass filtered at 2 kHz, digitally
sampled at 5 kHz, and analyzed with pClamp software version 9.2 (Molecular Devices). Patch
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
+
pipettes (3–5 Mꢁ) pulled from borosilicate glass were filled with a Cs ꢀbased intracellular
solution containing the following (in mM): 120 gluconic acid δ-lactone, 15 CsCl, 10 BAPTA,
1
0 HEPES, 3 MgCl , 1 CaCl and 2 ATPꢀMg salt (pHꢀadjusted to 7.2 with CsOH). Extracellular
2 2
solution (ECS) contained the following (in mM): 160 NaCl, 2.5 KCl, 10 HEPES, 10 glucose, 0.7
CaCl and 0.2 EDTA (pHꢀadjusted to 7.3 with NaOH). NMDAR responses were induced by
2
1
ꢁM or 1 mM glutamate (in recombinant receptors) and 10 ꢁM NMDA (native receptors)
together with 10 ꢁM glycine. The ECS used for native receptors had the same composition as the
ECS used for recombinant receptors. For native NMDAR recordings, ECS additionally
contained 10 ꢁM CNQX, 10 ꢁM bicuculline and 0.5 ꢁM TTX. AMPAR currents were induced
with 5 ꢁM AMPA and the ECS additionally contained 50 ꢁM DꢀAP5, 10 ꢁM bicuculline, 0.5 ꢁM
TTX and 10 ꢁM cyclothiazide. GABA R responses were induced with 5 ꢁM GABA and the
A
ECS additionally contained 50 ꢁM DꢀAP5, 10 ꢁM CNQX and 0.5 ꢁM TTX. Compound
solutions were prepared fresh before each experiment as a stock solution in dimethyl sulfoxide
(DMSO). The same concentration of DMSO (1%) was maintained in all test and control
extracellular solutions. Experiments were performed at room temperature (21–25 °C).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 48
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Experimental Data for Compounds 1ꢀ28
General procedure I: Synthesis of steroidal hemiesters from anhydrides
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
A mixture of steroid 25, 26, 27, or 28 (1 mmol) and dicarboxylic acid anhydride (6 equiv) was
dried overnight at 50 °C. Dry pyridine (12 mL) and DMAP (2.4 equiv) were added. The mixture
was heated at 110 °C for 6 h under an inert atmosphere. It was then poured into water and
extracted with DCM. Combined organic extracts were washed with brine and dried. Solvents
were evaporated and the residue was purified on a column of silica gel.
General procedure II: Synthesis of steroidal hemiesters from ωꢀdicarboxylic acids
A solution of steroid 25, 26, 27, or 28 (1 mmol) in dry DCM (5 mL) was added under an inert
atmosphere to a solution of dicarboxylic acid (2 mmol), EDCI (2 mmol), DIPEA (2 mmol), and
DMAP (2.4 equiv) in dry DCM (10 mL) at 0 °C under stirring. The reaction mixture was stirred
at room temperature for 18 h. The solvents were then evaporated and the residue was dissolved
with DCM, washed with water, dried over sodium sulfate, and concentrated in vacuo.
Purification on a column of silica gel gave the desired steroidal hemiester.
2
0ꢀOxoꢀpregnꢀ5ꢀenꢀ3βꢀyl sulfate Pyridinium Salt (1, PES). Sulfur trioxide pyridine complex
(318 mg, 2 mmol) was added to a solution of compound 25 (316 mg, 1 mmol) in chloroform (5
mL, P O dried and distilled) and dry pyridine (1 drop). The reaction mixture was stirred at room
2
5
temperature under an inert atmosphere for 6 h. The progress of the reaction was checked by
TLC. The reaction mixture was then allowed to stand at −18 °C for 2 h and the solids were
filtered off through a plug of cotton wool. The filtrate was evaporated and dried (25 °C, 100 Pa)
ACS Paragon Plus Environment

Page 27 of 48
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
for 1 h. The crude material was dissolved in freshly dried chloroform (4 mL), allowed to stand at
−18 °C for 2 h, and the solid was filtered off. The filtrate was evaporated, yielding compound 1
2
0
1
(425 mg, 89%): m.p. 182–184 °C (CHCl ); [α]D 18.1 (c 0.3, CHCl ). H NMR (400 MHz,
3
3
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
CHCl ): δ 0.64 (3H, s, Hꢀ18), 1.02 (3H, s, Hꢀ19), 2.14 (3H, s, Hꢀ21), 2.55 (1H, t, J = 8.9, Hꢀ17),
3
2
.65 (1H, ddd, J = 2.3, J = 5.1, J = 13.3, Hꢀ4), 4.37 (1H, tt, J = 4.8, J = 11.6, Hꢀ3), 5.40 (1H,
1 2 3 1 2
d, J = 5.2, Hꢀ6), 8.02 (2H, m, Hꢀ2′ and Hꢀ4′, pyridinium), 8.49 (1H, m, Hꢀ3′, pyridinium), 9.00
1
3
(
2H, m, Hꢀ1′ and Hꢀ5′, pyridinium). C NMR (101 MHz, CDCl ): δ 209.7 (C=O), 145.7 (Cꢀ1′
3
and Cꢀ5′, pyridinium), 142.5 (Cꢀ3′, pyridinium), 140.3 (Cꢀ5), 127.2 (Cꢀ2′ and Cꢀ4′, pyridinium),
22.2 (Cꢀ6), 78.9 (Cꢀ3), 63.8, 57.0, 50.0, 43.9, 39.3, 38.9, 37.3, 36.6, 31.9, 31.9, 31.7, 28.9, 24.6,
22.9, 21.2, 19.4, 13.3. IR (CHCl ): 1698 (C=O); 1665 (C=C); 1490 (pyridinium); 971, 952
1
3
(
COS). MS: ESI m/z 395.3 (100%, M – pyH). HRꢀMS (ESI) m/z: for C H O S [M – pyH]
21 31 5
calcd, 395.18977; found, 395.19906.
9
a
2
0ꢀOxoꢀpregnꢀ5ꢀenꢀ3βꢀyl hemioxalate (2, 20ꢀOxoꢀPEꢀhOxa). A mixture of compound 25
(316 mg, 1 mmol), dry DCM (5 mL), triethylamine (0.15 mL), and two drops of previously
prepared solution (2 mL of dry DCMDCM and 1 drop DMF) was added to a cooled (0 °C)
mixture of DCM (5 mL) and oxalyl chloride (0.26 mL, 3 mmol). The mixture was allowed to
reach 10 °C and then stirred for 2 h under these conditions. 20 mL of water was then added to
decompose the excess reagent, and the mixture was stirred for 30 min at room temperature. The
organic layer was separated and ethyl acetate (20 mL) was added to the DCM extract. An
aqueous solution of potassium carbonate was then added (10%, 50 mL), the organic layer with
undesired byꢀproducts was separated, and then an aqueous solution of HCl was cautiously added
to the aqueous layer (1N, to pH ~4). The crude product was obtained by extraction with ethyl
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 48
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
acetate (2 x 25 mL) and dried. Chromatography on a silica gel column (5% methanol in DCM)
20
1
gave 2 (140 mg, 36%): m.p. 189–191 °C (ethyl acetate/nꢀheptane); [α]D 6.2 (c 0.3, CHCl ). H
3
NMR (400 MHz, CDCl ): δ 0.64 (3H, s, Hꢀ18), 1.04 (3H, s, Hꢀ19), 2.14 (3H, s, Hꢀ21), 2.54 (1H,
3
1
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
t, J = 8.9, Hꢀ17), 4.81 (1H, m, Hꢀ3), 5.43 (1H, m, Hꢀ6). C NMR (101 MHz, CDCl ): δ 210.4
3
(C=O), 157.9 (COOH), 157.7 (COO), 138.8 (Cꢀ5), 123.4 (Cꢀ6), 78.3 (Cꢀ3), 63.8 (Cꢀ17), 56.9,
4
9.9, 44.2, 38.8, 37.6, 36.9, 36.7, 31.9, 31.8, 31.7, 27.4, 24.6, 22.9, 21.1, 19.3, 13.3. IR (CHCl3):
1
801, 1760, 1699 (C=O); 1672 (C=C); 1192 (CꢀO). MS: ESI m/z 387.2 (100%, M – 1), 388.2
(
24%, M). HRꢀMS (ESI) m/z: for C H O [M – 1] calcd, 387.21660; found, 387.21676.
23 31 5
20ꢀOxoꢀpregnꢀ5ꢀenꢀ3βꢀyl hemimalonate (3, 20ꢀOxoꢀPEꢀhMal). Dry toluene (15 mL) was
added to a mixture of compound 25 (316 mg, 1 mmol) and 2,2ꢀdimethylꢀ4,6ꢀdioxoꢀ1,3ꢀdioxolane
Meldrum´s acid, 171 mg, 1.5 mmol). The mixture was heated at 80 °C for 48 h under an inert
(
atmosphere. It was then washed with brine and the solvent was evaporated. Chromatography of
the residue on a silica gel column (30ꢀ50% ethyl acetate in petroleum ether) gave 3 (149 mg,
20
1
3
0
7%): m.p. 166–168 °C (acetone); [α]D 10.6 (c 0.2, CHCl ). H NMR (400 MHz, CDCl ): δ
3 3
.63 (3H, s, Hꢀ18), 1.02 (3H, s, Hꢀ19), 2.13 (3H, s, Hꢀ21), 2.54 (1H, t, J = 8.8, Hꢀ17), 3.42 (2H,
1
3
s, Hꢀ2´), 4.71 (1H, m, Hꢀ3), 5.39 (1H, m, Hꢀ6). C NMR (101 MHz, CDCl ): δ 209.9 (C=O),
3
1
4
69.5 (COOH), 167.4 (COO), 139.3 (Cꢀ5), 122.9 (Cꢀ5), 76.0 (Cꢀ3), 63.8 (Cꢀ17), 56.9, 49.9, 44.1,
0.4, 38.9, 37.9, 37.0, 36.7, 31.9, 31.9, 31.7, 27.6, 24.6, 22.9, 21.1, 19.4, 13.3. IR (CHCl ): 1746,
3
1736, 1725, 1706 (C=O); 1675 (C=C); 1192 (CꢀO). MS: ESI m/z 357.3 (100%, M – 1 – CO2),
4
01.3 (10%, M – 1). HRꢀMS (ESI) m/z: for C H O [M – 1] calcd, 401.23335; found,
24 33 5
4
01.23291.
ACS Paragon Plus Environment

Page 29 of 48
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
9a
2
0ꢀOxoꢀpregnꢀ5ꢀenꢀ3βꢀyl hemisuccinate (4, 20ꢀOxoꢀPEꢀhSuc). Compound 4 was prepared
according to General procedure I. Starting from compound 25 (251 mg, 0.79 mmol), compound
4 (172 mg, 52%) was obtained as a white solid by column chromatography (30ꢀ50% ethyl
20
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
acetate in petroleum ether): m.p. 161–163 °C (toluene); [α]D +14.3 (c 0.2, CHCl ). H NMR
3
(
400 MHz, CDCl ): δ 0.63 (s, 3H, Hꢀ18), 1.02 (3H, s, Hꢀ19), 2.12 (3H, s, Hꢀ21), 2.61–2.68 (4H,
3
1
3
m, OOCCH CH COO), 4.62 (1H, m, Hꢀ3), 5.37 (1H, m, Hꢀ6). C NMR (101 MHz, CDCl ): δ
2
2
3
2
5
09.8 (C=O), 177.0 (COOH), 171.7 (COO), 139.6 (Cꢀ5), 122.5 (Cꢀ6), 74.5 (Cꢀ3), 63.8, 56.9,
0.0, 44.1, 38.9, 38.1, 37.1, 36.7, 31.9, 31.9, 31.7, 29.3, 28.9, 27.8, 24.6, 22.9, 21.1, 19.4, 13.3.
IR (CHCl ): 1731, 1716 (C=O); 1671 (C=C). MS: ESI m/z 415.3 (100%, M – 1). For C H O
3
25 36
5
(416.5) calcd: 72.08%, C; 8.71%, H. Found: 71.85%, C; 8.63%, H.
9
a
2
0ꢀOxoꢀpregnꢀ5ꢀenꢀ3βꢀyl hemiglutarate (5, 20ꢀOxoꢀPEꢀhGlu). Compound 5 was prepared
according to General procedure I. Starting from compound 25 (316 mg, 1 mmol), compound 5
219 mg, 51%) was obtained as a white solid by column chromatography (2ꢀ5% acetone in
(
20
1
DCM): m.p. 137–139 °C (methanol); [α]D 9.1 (c 0.3, CHCl ). H NMR (400 MHz, CDCl ): δ
3
3
0
.63 (3H, s, Hꢀ18), 1.02 (3H, s, Hꢀ19), 1.96 (2H, p, J = 7.3, Hꢀ3´), 2.12 (3H, s, Hꢀ21), 2.37 (2H,
t, J = 7.3, Hꢀ2´), 2.43 (2H, t, J = 7.3, Hꢀ4´), 2.54 (1H, t, J = 8.9, Hꢀ17), 4.63 (1H, m, Hꢀ3), 5.37
13
(
1H, m, Hꢀ6). C NMR (101 MHz, CDCl ): δ 209.8 (C=O), 177.9 (COOH), 172.4 (COO), 139.7
3
(
Cꢀ5), 122.5 (Cꢀ6), 74.1 (Cꢀ3), 63.8 (Cꢀ17), 56.9, 50.0, 44.1, 38.9, 38.2, 37.1, 36.7, 33.6, 32.9,
31.9, 31.9, 31.7, 27.8, 24.6, 22.9, 21.1, 20.0, 19.4, 13.3. IR (CHCl ): 1725, 1711 (C=O); 1655
3
(
C=C); 1417 (CꢀO). MS: ESI m/z 429.3 (100%, M – 1), 430.3 (30%, M). For C H O (430.3)
26 38 5
calcd: 72.53%, C; 8.90% H. Found: 72.36%, C; 8.89%, H.
ACS Paragon Plus Environment