Synthesis and biological evaluation of benzodiazepines containing a pentafluorosulfanyl group

Article abstract of DOI:10.1016/j.tet.2021.132020

The widely used pentafluorosulfanyl group (SF₅) was deployed as a bioisosteric replacement for a chloro-group in the benzodiazepine diazepam (Valium). Reaction of 2-amino-5-pentafluorosulfanyl-benzophenone with chloroacetyl chloride followed by hexamethylenetetramine, in the presence of ammonia, led to 7-sulfurpentafluoro-5-phenyl-1H-benzo[1,4]diazepin-2(3H)-one (2c). The latter was able to undergo a Pd-catalysed ortho-arylation, demonstrating that these highly fluorinated benzodiazepines can be further modified to form more complicated scaffolds. The replacement of Cl by the SF₅ group, led to a loss of potency for potentiating GABA_A receptor activation, most likely because of a lost ligand interaction with His102 in the GABA_A receptor α subunit. Dedicated to Professor Jonathan Williams, an inspirational and humble pioneer, a colleague and mentor in chemistry.

Full text of DOI:10.1016/j.tet.2021.132020

Tetrahedron 85 (2021) 132020
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Synthesis and biological evaluation of benzodiazepines containing a
pentafluorosulfanyl group
Arathy Jose a, Raysa Khan Tareque a, Martin Mortensen b, Remi Legay c, Simon J. Coles d,
,
**
Graham J. Tizzard d, Barnaby W. Greenland a, Trevor G. Smart b, Mark C. Bagley a
,
,
*
John Spencer a
a Chemistry Department, School of Life Sciences, Falmer, Brighton, BN1 9QJ, UK
b Department of Neuroscience, Physiology & Pharmacology, Division of Biosciences, University College London, Gower Street, London, WC1E 6BT, UK
Normandie Universite, Laboratoire de Chimie Moleculaire et Thioorganique LCMT UMR 6507, ENSICAEN, UNICAEN, CNRS, 6 Bd. Du Marechal Juin, 14050,
Caen, France
c
ꢀ
ꢀ
ꢀ
d National Crystallography Service, School of Chemistry, University of Southampton, Southampton, SO17 1BJ, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
The widely used pentafluorosulfanyl group (SF5) was deployed as a bioisosteric replacement for a chloro-
group in the benzodiazepine diazepam (Valium™). Reaction of 2-amino-5-pentafluorosulfanyl-benzo-
phenone with chloroacetyl chloride followed by hexamethylenetetramine, in the presence of ammonia,
led to 7-sulfurpentafluoro-5-phenyl-1H-benzo[1,4]diazepin-2(3H)-one (2c). The latter was able to un-
dergo a Pd-catalysed ortho-arylation, demonstrating that these highly fluorinated benzodiazepines can
be further modified to form more complicated scaffolds. The replacement of Cl by the SF5 group, led to a
loss of potency for potentiating GABAA receptor activation, most likely because of a lost ligand interaction
Received 23 January 2021
Received in revised form
5 February 2021
Accepted 8 February 2021
Available online 20 February 2021
Keywords:
with His102 in the GABAA receptor
a subunit.
Benzodiazepines
Medicinal chemistry
GABA
Dedicated to Professor Jonathan Williams, an inspirational and humble pioneer, a colleague and
mentor in chemistry.
© 2021 Elsevier Ltd. All rights reserved.
Bioisosteres
Electrophysiology
HEK cells
1. Introduction
With research outputs in both benzodiazepine [16,17] and SF5
chemistry [18], there was a natural inclination for us to combine
The pentafluorosulfanyl group is often employed in medicinal
chemistry as a bioisosteric “super trifluoromethyl” group. Pos-
sessing high thermal stability, low toxicity, electron withdrawing
effects and high lipophilicity, it has been used in a number of drug
discovery projects [1e6].
these interests in the design of SF5-containing BZDs. We, therefore,
aimed to synthesise analogues 2a - 2c (Scheme 1) related to the
much-prescribed drug, diazepam (Valium™) in order to evaluate
the effect of changing a Cl for a SF5 group on biological activity.
Since their discovery in the late 1950s, benzodiazepines (BZDs)
which act as positive allosteric modulators on g-subunit containing
2. Results and discussion
synaptic GABAA receptors (GABAARs) [7,8], have been widely
employed to treat a wide spectrum of disorders such as anxiety,
insomnia, seizures and alcohol withdrawal [9e12]. Structure ac-
tivity relationships show, inter alia, that electron withdrawing
groups at the 7-position are important for improved receptor af-
finity (Fig. 1) [13e15].
We opted for a one-pot microwave route to synthesise SF5-
substituted BZD analogues. [19e21] Commercially available 2-
amino-5-pentafluorosulfanyl-benzophenone 1c was coupled un-
der microwave irradiation with Boc-Gly-OH, and DCC as the
coupling agent, in toluene at 150 ꢀC for 30 min, followed by Boc-
deprotection with TFA [17,22]. However, the attempt was unsuc-
cessful and one speculation for the failure was the poor nucleo-
philicity of the aniline. To validate this hypothesis, we attempted
the same reaction with 2-amino-5-nitrobenzophenone as the nitro
group has an electronic effect fairly close to that of the SF5 group
* Corresponding author.
** Corresponding author.
E-mail address: j.spencer@sussex.ac.uk (J. Spencer).
https://doi.org/10.1016/j.tet.2021.132020
0040-4020/© 2021 Elsevier Ltd. All rights reserved.

A. Jose, R.K. Tareque, M. Mortensen et al.
Tetrahedron 85 (2021) 132020
this initial coupling step was responsible, or the cyclisation step, for
the poor overall yield. We found that the coupling step was very
low yielding for the reaction of 1b and the reaction was also,
disappointingly, again, unsuccessful for 1c.
As the microwave-mediated attempts towards the SF5-BZD de-
rivatives were unsuccessful, yet worked on a standard 1,4-BZD core
(entry 1), we sought a route towards the desired products using
other protocols. A method using hexamethylenetetramine [24e26]
with ammonia as aminating reagent, was reported to be successful,
even for starting materials with electron-withdrawing sub-
stituents. Accordingly, this was our next method of choice.
Hence, 1c was acylated using chloroacetyl chloride then ami-
nated with hexamethylenetetramine in the presence of ammonia
(Scheme 2). Analysis of the crude mixture, gratifyingly, showed the
presence of the expected product as well as a similar by-product,
which we tentatively assigned the structure 4c, notably by the
similarity of its 1H NMR [26] spectrum to that of its 4-chloro-de-
rivative. The two products could be separated after a normal phase
and a reverse phase column chromatographic purification.
Compound 2c was crystallised by a diffusion method using
dichloromethane/hexane and obtained as colourless crystals and
this confirmed both the regiochemistry of the SF5-substituent and
the formation of the BZD core (CCDC Deposition Number 2064966)
(Scheme 2) [27].
A standard N-methylation of 2c using sodium hydride and
methyl iodide yielded the desired SF5-BZD 5c in modest yield (see
Scheme 3).
An unoptimised attempt at Pd-catalysed CeH activation with
iodonium salts [28], using our previously described conditions,
involving microwave chemistry afforded the expected ortho-
aryated product 6c (see Scheme 4). This illustrates that catalytic
CeH activation chemistry is now amenable to the synthesis of
polyfluorinated BZDs and 6c was now available for biological assay.
The compounds were docked into the cryo-EM structure (PDB
Fig. 1. Selected clinically-used BZDs.
Scheme 1. Synthesis of BZDs by microwave techniques.
(
sp ¼ 0.68 [5] for SF5 and sp ¼ 0.78 [23] for NO2). The result was as
postulated, unsuccessful.
Although position-8 on the BZD ring was not a region of interest
in terms of biological activity, we were curious about the electronic
effect a pentafluorosulfanyl group would lead to at this position.
Again, we used the microwave approach for the attempted syn-
thesis of 2b (Scheme 1).
The reaction was moderately successful with 2b formed in 13%
yield with only a purity of 88% by LCMS. The unsubstituted
benzodiazepine 2a was synthesized in 65% yield.
Unperturbed in this approach, we next attempted the micro-
wave mediated route, utilizing 1 and Boc-Gly-OH but with EEDQ as
the coupling agent (Table 1). Moreover, the coupling reaction
mixture was worked up and the anticipated intermediate was
isolated and purified before continuing to the next step, viz. Boc-
group deprotection. This would enable us to establish whether
ID: 6HUP) of the
a1b3g2L GABAA receptor at the interfacial
benzodiazepine binding site between the principal (þ)
a and
€
complementary (ꢁ)
g
subunit using Schrodinger Glide [29]. We
evaluated their apparent binding affinity using the Glide score,
which predicts possible binding of the ligands in the benzodiaze-
pine binding site of the receptor and produces a set of initial ligand
conformations. Different ligand poses can then be generated and
ranked. Scoring is related to the strength of interaction between the
ligand and the protein which is expressed as binding free energy
[30]. Therefore, more negative values represent tighter binders.
Glide is primarily concerned with generating accurate poses for
each protein-ligand complex and identifying poses with appre-
ciable binding affinity. However, the task of accurately estimating
Table 1
Boc-Gly-OH coupling reactions.
Entry
1
Product
Yield (%)
86
2
3
10
e
Scheme 2. Multi-step synthesis of a SF5 substituted BZD.
2

A. Jose, R.K. Tareque, M. Mortensen et al.
Tetrahedron 85 (2021) 132020
Scheme 3. N-Methylation of a SF5-BZD.
Scheme 4. Ortho-arylation of a BZD.
protein ligand binding affinities is beyond the capabilities of
docking scoring functions and, hence Glide scores are not always
congruent with experimental data [31].
A Glide score was determined for compounds 2c, 5c and 6c and
was compared against diazepam and the metabolite, nordiazepam
(Table 2). The SF5-substituted nordiazepam analogue, 2c, however
gave a better Glide score than diazepam suggesting it may bind
more strongly in the binding site. The Glide score of the ortho CeH
activated analogue 6c was very poor in comparison.
Fig. 2. Benzodiazepines in complex with
a
1
b
3
g
2L receptor a) Brown dashed lines
show hydrophobic clashes between Diazepam’s (purple) chlorine and amino acid
residues, His102, Val203, and Tyr210. b) Diazepam (purple) overlapped with SF5-dia-
zapam (5c, red). c) Interactions between nordiazepam (blue) and the benzodiazepine
binding pocket and d) superposition of nordiazepam and SF5-nordiazapam (2c, pink).
e) 6c (Violet) in complex with a1b3g2L receptor shows no interaction with His102.
Aromatic hydrogen bonds are indicated by blue dashed lines. Yellow dashed lines
indicate hydrogen bonds. Pi-pi interactions are indicated by blue dashed lines. SF5-
Poses of diazepam, 2c, 5c and 6c respectively with the best Glide
score docked in the
dashed lines indicate hydrogen bonds and
chlorine atom interacts with the critical His102 side chain [32,33].
The distance between the chlorine atom and the nitrogen on the
a
1b
3
g2L receptor are shown in Fig. 2. The
p-p
interactions. The
a
a
His102 was measured as 2.89 Å.
The images show that there is no interaction between SF5 and
diazapam and SF5-nordiazapam do not interact with the His102 which is
interaction between Diazepam and the binding pocket.
a key
the amino acid side chains. A direct comparison of 5c and diazepam
can be made. We calculated the distance between SF5 and His102
to be 5.34 Å. This was calculated between the closest fluorine of SF5
to His102. This distance is almost double the distance between
chlorine and His102 for Diazepam (2.89 Å). This could explain the
lack of interaction between SF5 and the His102. This increased
a
a
a
a
distance and lack of interaction also applies to 2c and 6c as well.
To access functionality of the BZD ligands, we used whole-cell
patch-clamp recording from human embryonic kidney cells
expressing recombinant
and 6c were compared to diazepam for their ability to potentiate
M GABA-induced currents (~EC6.5). The three SF5-diazepam
a1b2g2L GABAARs. The analogues, 2c, 5c
2
m
analogues showed much lower potencies than diazepam (shifted
60- (2c), 70- (6c), and 190-fold lower (5c)). The relative extent of
potentiation was very low for 6c, ~half that of diazepam for 5c, or
near equivalent with diazepam for 2c (Fig. 3; Table 3). For 6c, the
efficacy level of the potentiation was reduced at the highest con-
centration of 100 mM (Fig. 3). Such inhibition has been reported
before for benzodiazepines like diazepam and flurazepam [34], and
Fig. 3. Concentration-response curves for the potentiation of GABA-induced currents
by diazepam (black), 2c (green), 5c (red), and 6c (blue) on recombinant 2L
a1b2g
GABAARs expressed in HEK293 cells. Data points (mean SEM; n ¼ 5) are plotted as
Table 2
percentages of current potentiation above that induced by 2
of modulator.
mM GABA in the absence
Glide score of Diazepam versus SF5-substituted BZDs.
Entry
1,4-BZD
Glide score
1
2
3
4
5
diazepam
nordiazepam
2c
5c
6c
ꢁ4.74
ꢁ5.36
ꢁ4.97
ꢁ4.74
ꢁ3.98
could reflect increased desensitization of GABAARs.
From these data, it is clear that substituting Cl on the benzo ring
for SF5 has a deleterious effect primarily on BZD potency and to a
3

A. Jose, R.K. Tareque, M. Mortensen et al.
Tetrahedron 85 (2021) 132020
Table 3
Mean maximum potentiation and potency values for diazepam and SF5-substituted BZDs for modulating GABAA receptors.
Diazepam
2c
5c
6c
Maximum Potentiation
133 15%
138 19%
77 20%
28 2.5%
Potency pEC50 SEM (EC50
)
7.52 0.07 (30 nM)
5.78 0.04 (1.7
m
M)
5.25 0.12 (5.7
m
M)
5.70 0.08 (2.0 mM)
large extent, also on relative efficacy at GABAA receptors. This is
likely to be due to disruption of the Cl e H102 interaction, which is
known to be critical for BZD modulation at GABAA receptors.
Indeed, mutation of His for Arg at this location, and found in BDZ-
insensitive a4 and a6 receptors, completely abolishes BZD modu-
lation of GABAA receptor activation [32,34].
added and the reaction mixture was refluxed overnight. After
cooling, the latter was concentrated in vacuo and dissolved in
toluene (5 mL). p-Toluene sulfonic acid (6 mg, 0.03 mmol), was
added to the solution and the mixture was refluxed for 1 h. The
crude was concentrated in vacuo and purified over a column of
silica (hexane:EtOAc; 3:7), followed by a reverse phased column
(C18, acetonitrile:water, 1:3) to obtain the pure product as a col-
a
ourless solid (83 mg, 31%). 1H NMR (600 MHz, Chloroform-d)
d 9.40
3. Conclusion
(s, 1H, NH), 7.87 (dd, J ¼ 8.9, 2.6 Hz, 1H, ArH), 7.74 (d, J ¼ 2.6 Hz, 1H,
ArH), 7.53e7.50 (m, 2H, 2ArH), 7.50e7.47 (m, 1H, ArH), 7.41 (pt, 2H,
2ArH), 7.25 (d, J ¼ 8.9 Hz, 1H, ArH), 4.37 (s, 2H, CH2); 13C NMR
Selected SF5-substituted 1,4-BZDs have been synthesized, one
by a Pd-catalysed CeH activation method, and evaluated in silico
and in vitro for their biological activity. For all compounds, which
are direct analogues of diazepam, where a Cl has been replaced by a
SF5 group, reduced GABA potency and for 5c and 6c a reduced ef-
ficacy were evident.
(600 MHz, Chloroform-d)
d 172.1 (C]O), 169.8 (C]N), 148.1 (t,
1JF,C ¼ 18.9 Hz, ArC-SF5), 141.0 (ArC), 138.2 (ArC), 131.1 (ArC), 129.6
(2ArC), 129.5 (m, ArC), 129.1e129.01 (m, ArC), 128.5 (2ArC), 126.8
(ArC), 121.5 (ArC), 56.7 (CH2); 19F NMR (376 MHz, Chloroform-d)
d
83.51 (q, J ¼ 150.5 Hz), 63.32 (d, J ¼ 150.5 Hz); LCMS Purity
(UV) ¼ 96%, tR 18.11 min; HRMS - ESI (m/z) found 385.0404, calc. for
[C15H11F5N2OS] [þNa]þ: 385.0404; IR (neat) nmax/cmꢁ1: 3089
(NeH), 1688 (C]O), 1610 (C]N), 824 (SeF); mp ¼ 158e159 ꢀC.
4. Experimental
3-Amino-6-(pentafluoro-l
6-sulfanyl)-4-phenyl-1H-quinolin-
4.1. Organic chemistry
2-one was isolated from the reversed phase column of 2c as a
colourless solid (33 mg, 12%, <90% purity by LCMS after several
All commercially purchased materials and solvents were used
more attempted purifications) 1H NMR (600 MHz, CD3CN)
d 10.28
without further purification unless specified otherwise. NMR
spectra were recorded on a Varian VNMRS 600 (1H 600 MHz, 13
C
(s, 1H), 7.65e7.61 (m, 3H), 7.56e7.53 (m, 1H), 7.39e7.35 (m, 3H),
7.31 (d, J ¼ 2.5 Hz, 1H), 4.69 (s, 2H).
126 MHz) and VNMRS 400 (19F 376 MHz, 2H 61 MHz and 31P
162 MHz) spectrometer and prepared in deuterated solvents such
as Chloroform-d and DMSO-d6. 1H and 13C chemical shifts were
recorded in parts per million (ppm). Multiplicity of 1H NMR peaks
are indicated by s e singlet, d e doublet, dd e doublets of doublets,
t e triplet, pt e pseudo triplet, q e quartet, m e multiplet and
coupling constants are given in Hertz (Hz). Electronspray ionisation
e high resolution mass spectra (ESI-HRMS) were obtained using a
Bruker Daltonics Apex III where Apollo ESI was used as the ESI
source. All analyses were conducted by Dr A. K. Abdul-Sada. The
molecular ion peaks [M]þ were recorded in mass to charge (m/z)
ratio. LC-MS spectra were acquired using Shimadzu LC-MS 2020, on
a Gemini 5 m C18 110 Å. column. X-ray analysis was performed at
the UK National Crystallography Services, Southampton. Purifica-
tions were performed by flash chromatography on silica gel col-
umns or C18 columns using a Combi flash RF 75 PSI, ISCO unit.
tert-Butyl-N-({[2-benzoyl)
phenyl]carbamoyl}
methyl)
carbamate (3a) [20]. 2-Aminobenzophenone (300.0 mg,
1.52 mmol), EEDQ (376.0 mg, 1.52 mmol), Boc-Gly-OH (268.0 mg,
1.52 mmol) and DCM (3 mL) were subjected to microwave irradi-
ation at 150 ꢀC for 30 min at 200 W. After 30 min, the reaction
mixture was diluted with DCM (5 mL) and washed with 10% HCl
(3 ꢂ 5 mL). The organic layer was extracted with DCM (2 x 5 mL),
dried over MgSO4, filtered and concentrated in vacuo. The crude
was purified over a column of silica (hexane: EtOAc; 7:3) to obtain
the title compound as a colourless solid (465 mg, 86%). 1H NMR
(600 MHz, dmso-d6)
d
10.51 (s, 1H), 7.54 (t, J ¼ 7.7 Hz, 1H), 7.46 (dd,
J ¼ 17.5, 7.7 Hz, 3H), 7.40 (t, J ¼ 7.5 Hz, 2H), 7.24e7.19 (m, 2H), 7.15 (t,
J ¼ 7.5 Hz, 1H), 4.08 (s, 2H), 2.47 (s, 9H). Known compound.
tert-Butyl-N-({[2-benzoyl-4-(pentafluoro-l⁶-sulfanyl)
phenyl]carbamoyl} methyl)carbamate (3b). 2-benzoyl-4-(penta-
fluorosulfanyl)aniline methanone (100.0 mg, 0.31 mmol), EEDQ
(77.0 mg, 0.31 mmol), Boc-Gly-OH (55.0 mg, 0.31 mmol) and DCM
(1 mL) were subjected to microwave irradiation at 150 ꢀC for
30 min at 200 W. After 30 min, the reaction mixture was diluted
with DCM (5 mL) and washed with 10% HCl (3 ꢂ 5 mL). The organic
layer was extracted with DCM (2 x 5 mL), dried over MgSO4, filtered
and concentrated in vacuo. The crude was purified over a column of
silica (Hexane: EtOAc; 7:3) to obtain the title compound as a col-
ourless solid. (15 mg, 10%). 1H NMR (600 MHz, Chloroform-d)
7-(Pentafluoro-
benzodiazepin-2-one (2c). Triethylamine (0.188 g, 1.86 mmol) was
added to a solution of 2-benzoyl-4-(pentafluoro-
6-sulfanyl)aniline
l
6-sulfanyl)-5-phenyl-2,3-dihydro-1H-1,4-
l
(0.300 g, 0.93 mmol) in dichloromethane (1 mL) and the mixture
was stirred for 1 h at room temperature. After an hour the mixture
was cooled in an ice bath, and chloroacetyl chloride (0.210 g,
1.86 mmol) dissolved in dichloromethane (1 mL) and cooled in an
ice bath was added dropwise to the reaction mixture. The reaction
was stirred overnight at room temperature. The reaction was
monitored by TLC and the crude was concentrated on vacuo and
purified by flash chromatography (petroleum ether: ethyl acetate;
d
11.12 (s, 1H, NH), 9.18 (s, 1H), 7.71e7.68 (m, 2H, ArH), 7.65e7.62
(m, 2H, ArH), 7.51 (d, J ¼ 7.8 Hz, 2H, ArH), 7.49e7.46 (m, 2H, ArH),
3.99 (d, J ¼ 6.0 Hz, 2H, ArH), 1.44 (s, 9H, (CH3)3). Insufficient ma-
terial for 13C spectrum.
7:3) to obtain pure N-[2-benzoyl-4-(pentafluoro-l
6-sulfanyl)
phenyl]-2-chloroacetamide as a colourless solid (301 mg, 81%).
Ammonium carbonate (0.360 g, 3.75 mmol) was suspended in a
solution of ammonia (2 M) in ethanol (5 mL) and stirred. Hexa-
methylenetetramine (0.531 g, 3.75 mmol) was added and refluxed.
After 5 min of refluxing, a solution of N-(2-benzoyl-4-sulfur penta-
fluoro phenyl)-2-chloroacetamide in dichloromethane (3 mL) was
1-methyl-7-(pentafluoro-l
6-sulfanyl)-5-phenyl-2,3-dihydro-
1H-1,4-benzodiazepin-2-one (5c). Sodium hydride (0.023 g,
0.94 mmol) was added to a solution of 2c (0.170 g, 0.47 mmol) in
dry THF (1 mL) and the mixture was stirred for 1 h. Methyl iodide
4

A. Jose, R.K. Tareque, M. Mortensen et al.
Tetrahedron 85 (2021) 132020
(0.133 g, 0.94 mmol) was added to the reaction mixture, which was
stirred at room temperature overnight. Crude product was washed
with water (3 ꢂ 15 mL), extracted with ethyl acetate (3 ꢂ 15 mL),
dried over MgSO4, filtered and concentrated in vacuo. The crude
was purified over a column of silica (hexane:EtOAc; 7:3) to obtain a
colourless solid as the title compound (66 mg, 37%). 1H NMR
4.4. Electrophysiology experiments
Whole-cell patch clamp recording from HEK cells was used to
study GABAA receptor currents as described previously [35] using
an Axopatch 200B Axon Instruments amplifier. Patch pipettes
(resistance 3e5 MU) were filled with a solution containing (mM):
(600 MHz, Chloroform-d)
d
7.93 (dd, J ¼ 9.1 Hz, 2.6 Hz, 1H, ArH),
120 KCl, 1 MgCl2, 11 EGTA, 30 KOH, 10 HEPES, 1 CaCl2, and 2
adenosine triphosphate; pH 7.11. The cells were continuously
perfused with Krebs recording solution containing (mM): 140 NaCl,
4.7 KCl, 1.2 MgCl2, 2.52 CaCl2, 11 Glucose and 5 HEPES; pH 7.4.
Diazepam and SF5-analogues were first dissolved in DMSO (stock),
and for functional electrophysiology experiments subsequently
diluted at least 1000-fold in Krebs solution. Drug solutions were
applied to recording cells via a Y-tube application system [30].
The potentiating effects of diazepam, and analogues 2c, 5c and
7.73 (d, J ¼ 2.6 Hz, 1H, ArH), 7.61 (m, 2H, ArH), 7.52 (m, 1H, ArH),
7.45 (m, 3H, ArH), 4.90 (d, J ¼ 11.0 Hz, 1H, CH), 3.79 (d, J ¼ 11.0 Hz,
1H, CH), 3.44 (s, 3H, CH3); 13C NMR (600 MHz, Chloroform-d)
d
169.8 (C]O), 168.8 (C]N), 148.4 (C-SF5), 146.1(ArC), 137.7 (ArC),
131.1 (ArC), 129.4 (2ArC), 128.7 (4ArC), 128.6(ArC), 121.2 (ArC), 56.9
(CH2), 34.9 (CH3); 19F NMR (400 MHz, Chloroform-d)
83.10 (p,
d
J ¼ 150.4 Hz), 63.23 (d, J ¼ 150.4 Hz); LCMS Purity (UV) ¼ 96%, tR
19.52 min; HRMS -ESI(m/z) found 377.0749, calc. for
[C16H13F5N2OS][þH]þ:377.0742; IR (neat) nmax/cmꢁ1: 1682 (C]O),
1611 (C]N), 835 (SeF); mp ¼ 240e242 ꢀC.
6c were evaluated in the presence of 2
mM GABA which was
equivalent to a current approximately 6.5% of the GABA maximum
response (EC6.5). The efficacy and potency for the potentiation by
each ligand was established by fitting curves to the GABA current
response-concentration relationship data points from each of the
five individual experiments using the Hill equation, I/Imax pot ¼ (1/
(1 þ (EC50/[L])n). The ligand potency, EC50, represents the concen-
tration of the ligand ([L]) inducing 50% of the maximal potentiation
5-{[1,1’-Biphenyl]-2-yl}-7-(pentafluoro-l
6-sulfanyl)-2,3-
dihydro-1H-1,4-benzodiazepin-2-one (6c). 2c (86 mg, 0.2
4 mmol), bis(2-fluorophenyl)iodonium tetrafluoroborate (145 mg,
0.36 mmol) and glacial acetic acid were combined in a 10 mL mi-
crowave vial. The vial was degassed and purged with argon before
adding palladium (II) acetate (7 mg, 0.0089 mmol, and 0.01equiv)
and stirring at 125 ꢀC in the microwave for 1 h. Thereafter, the
cooled reaction mixture was filtered through celite, washed with
dichloromethane (50 mL) and concentrated in vacuo. The residue
was dissolved in dichloromethane (15 mL), washed with sodium
bicarbonate (20 mL) and the organic layer extracted with
dichloromethane (20 mL x 3), dried over MgSO4, filtered and
concentrated in vacuo. The bright red oil was purified over a col-
umn of silica (hexane:EtOAc; 1:4) to obtain a colourless solid as
pure product (24 mg, 22%). 1H NMR (600 MHz, Chloroform-d)
current (in the presence of 2 mM GABA), and n is the Hill slope.
Since concentration response EC50 data are distributed on a
logarithmic scale, we converted these to pEC50 values (pEC50 ¼ -
log(EC50)) which are distributed on a linear scale. From pEC50
values we calculated mean values
sem, and to facilitate data-
interpretation we re-transformed these mean pEC50 values into
mean EC50 values (Table 3). The relative efficacy for GABA current
potentiation was calculated as a mean percentage
current induced by 2 M GABA alone.
sem of the
m
d
9.49 (s, 1H, NH), 7.82e7.74 (m, 1H, ArH), 7.58e7.53 (m, 2H,
2ArH), 7.50 (dd, J ¼ 8.9, J ¼ 2.5 Hz, 1H, ArH), 7.34e7.30 (m, 2H,
Declaration of competing interest
2ArH), 7.07e7.01 (m, 1H, ArH), 6.92e6.84 (m, 2H, ArH), 6.78 (d,
J ¼ 8.9 Hz, 1H, ArH), 6.75 (pt, J ¼ 9.0 Hz, 1H), 4.32 (s, 2H, CH2);13
C
The authors declare that they have no known competing
financial interests or personal relationships that could have
appeared to influence the work reported in this pape.
NMR (600 MHz, Chloroform-d)
d 171.1 (C]O), 170.9 (C]N), 158.7
(ArC-F, d, 1JF,C ¼ 247.6 Hz), 148.6e148.0(ArC-SF5), 139.5 (ArC), 138.9
(ArC), 135.7 (ArC), 131.3 (ArC), 131.0 (ArC),130.4 (ArC), 130.2 (ArC),
129.5 (ArC, d, 3JF,C ¼ 7.9 Hz), 128.5 (2ArC), 128.2 (ArC), 127.8 (ArC, d,
2JF,C ¼ 15.6 Hz), 127.6 (ArC), 123.7 (ArC), 120.5 (ArC), 115.0 (ArC, d,
2JF,C ¼ 22.0 Hz), 56.5 (CH2); 19F NMR (376 MHz, Chloroform-d)
Acknowledgements
The authors acknowledge financial support from the ERDF
(LabFact: InterReg V project 121). We thank Dr Alaa Abdul-Sada for
HRMS measurements.
d
83.40 (p, J ¼ 150.7 Hz, axial F), 63.07 (d, J ¼ 150.7 Hz, equatorial
F), ꢁ114.58 (F); LCMS Purity (UV) ¼ 98%, tR: 19.85min; HRMS
-ESI(m/z) found 457.0813, calc. for [C21H15F6N2OS][þH]þ:457.0809;
IR (neat) nmax/cmꢁ1: 3064 (NeH), 1704 (C]O), 1616 (C]N), 1336
(CeF), 835.5 (SeF); mp ¼ 190e191 ꢀC.
Appendix A. Supplementary data
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.tet.2021.132020.
4.2. Computational ligand docking
References
Docking was performed using the solved cryo-EM structure of
the
a1b3g2L receptor in complex with GABA and Diazepam ob-
[1] M.F. Sowaileh, R.A. Hazlitt, D.A. Colby, ChemMedChem 12 (2017) 1481e1490.
[2] J.M.W. Chan, J. Mater. Chem. C 7 (2019) 12822e12834.
[3] P. Beier, T. Pastýríkova, Beilstein J. Org. Chem. 9 (2013) 411e416.
[4] S. Mori, N. Tsuemoto, T. Kasagawa, E. Nakano, S. Fujii, H. Kagechika, Chem.
Pharm. Bull. 1278, 67, 1278e1283.
tained from PDB (ID: 6HUP). The software used was Schrodinger
Glide.
ꢁ
ꢀ
[5] P.R. Savoie, J.T. Welch, Chem. Rev. 115 (2015) 1130e1190.
[6] E. Pujol, N. Blanco-Cabra, E. Julian, R. Leiva, E. Torrents, S. Vazquez, Molecules
23 (2018) 1e17.
[7] T.I. Saari, M. Uusi-Oukari, J. Ahonen, K.T. Olkkola, Pharmacol. Rev. 63 (2011)
243e267.
[8] D.B. Pritchett, H. Sontheimer, B.D. Shivers, S. Ymer, H. Kettenmann,
P.R. Schofield, P.H. Seeburg, Nature 338 (1989) 582e585.
[9] S.C. Bell, T.S. Sulkowski, C. Gochman, S.J. Childress, J. Org. Chem. 27 (1962)
562e566.
4.3. Cell culture and recombinant GABAAR expression
ꢀ
ꢀ
HEK cells were maintained at 37 ꢀC, 95% CO2/5% O2 in Dulbecco’s
Modified Eagle’s Medium (DMEM) supplemented with 10% v/v fetal
bovine serum and 100 U/ml penicillin/100
Cells were transfected with cDNAs encoding enhanced green
fluorescent protein (EGFP) and murine 1, 2, 2L GABAAR subunits
mg/mL streptomycin.
a
b
g
[10] B. D, M. L, Eur. Psychiatr. 28 (2013) 7e20.
[11] W. Froestl, Future Med. Chem. 3 (2011) 163e175.
[12] L.H. Sternbach, Prog. Drug Res. 22 (1978) 229e266.
in a 1:1:1:1 ratio using a standard calcium-phosphate precipitation
method.
5

A. Jose, R.K. Tareque, M. Mortensen et al.
Tetrahedron 85 (2021) 132020
[13] N.E. Calcaterra, J.C. Barrow, ACS Chem. Neurosci. 5 (2014) 253e260.
161e176.
[14] J. Spencer, R.P. Rathnam, B.Z. Chowdhry, Future Med. Chem.
2
(2010)
[25] V.N. Anikeev, A.I. Petrunin, M.T. Kilin, F.V. Guss, Pharm. Chem. J. 38 (2004)
261e263.
1441e1449.
[15] N. Arora, P. Dhiman, S. Kumar, G. Singh, V. Monga, Bioorg. Chem. 97 (2020)
103668.
[16] J. Spencer, B.Z. Chowdhry, A.I. Mallet, R.P. Rathnam, T. Adatia, A. Bashall,
F. Rominger, Tetrahedron 64 (2008) 6082e6089.
[17] J. Spencer, R.P. Rathnam, A.L. Harvey, C.J. Clements, R.L. Clark, M.P. Barrett,
P.E. Wong, L. Male, S.J. Coles, S.P. MacKay, Bioorg. Med. Chem. 19 (2011)
1802e1815.
[18] S. Sansook, C.A. Ocasio, I.J. Day, G.J. Tizzard, S.J. Coles, O. Fedorov, J.M. Bennett,
J.M. Elkins, J. Spencer, Org. Biomol. Chem. 15 (2017) 8655.
[19] N. Kaur, D. Kishore, Synth. Commun. 44 (2014) 1375e1413.
[20] G. Mwande-Maguene, J. Jakhlal, J.B. Lekana-Douki, E. Mouray, T. Bousquet,
S. Pellegrini, P. Grellier, F.S.T. Ndouo, J. Lebibi, L. Pelinski, New J. Chem. 35
(2011) 2412e2415.
[26] US 4, Schering Corporation vol. 155, 1979, p. 904.
[27] S.J. Coles, P.A. Gale, Chem. Sci. 3 (2012) 683e689.
[28] R. Khan, S. Boonseng, P.D. Kemmitt, R. Felix, S.J. Coles, G.J. Tizzard, G. Williams,
O. Simmonds, J.-L. Harvey, J. Atack, H. Cox, J. Spencer, Adv. Synth. Catal. 359
(2017) 3261e3269.
€
[29] G. Schrodinger, LLC, N. York, NY, 2020.
[30] K. Raha, K.M. Merz, Annu. Rep. Comput. Chem. 1 (2005) 113e130.
[31] R.A. Friesner, J.L. Banks, R.B. Murphy, T.A. Halgren, J.J. Klicic, D.T. Mainz,
M.P. Repasky, E.H. Knoll, M. Shelley, J.K. Perry, D.E. Shaw, P. Francis,
P.S. Shenkin, J. Med. Chem. 47 (2004) 1739e1749.
[32] H.A. Wieland, H. Luddens, P.H. Seeburg, J. Biol. Chem. 267 (1992) 1426e1429.
ꢀ
[33] S. Masiulis, R. Desai, T. Uchanski, I. Serna Martin, D. Laverty, D. Karia,
T. Malinauskas, J. Zivanov, E. Pardon, A. Kotecha, J. Steyaert, K.W. Miller,
A.R. Aricescu, Nature 565 (2019) 454e459.
[34] R.J. Walters, S.H. Hadley, K.D.W. Morris, J. Amin, Nat. Neurosci. 3 (2000)
1274e1281.
[35] M. Mortensen, U. Kristiansen, B. Ebert, B. Frølund, P. Krogsgaard-Larsen,
T.G. Smart, J. Physiol. 557 (2004) 389e413.
[21] A. Liu, H. Zhou, G. Su, W. Zhang, B. Yan, J. Comb. Chem. 11 (2009) 1083e1093.
[22] J. Spencer, D.P. Sharratt, J. Dupont, A.L. Monteiro, V.I. Reis, M.P. Stracke,
F. Rominger, I.M. McDonald, Organometallics 24 (2005) 5665e5672.
[23] Y. Sumii, K. Sasaki, S. Tsuzuki, N. Shibata, Molecules 24 (2019) 3610.
ꢁ
ꢁ
ꢀ
ꢀ
ꢁ
[24] N. Blazevic, D. Kolbah, B. Belin, V. Sunjic, F. Kajfez, Synth. Met. 1979 (1979)
6