R. Giri, et al.
Bioorganic & Medicinal Chemistry Letters 30 (2020) 127305
Compounds 10a and 10c form docked poses with binding energies of
−
8.2 kcal/mol and −8.1 kcal/mol, respectively, which are very similar
to those in the D R binding pocket.
1
Overall, the computationally predicted binding energies for the
docked series of halogen substituted 1-phenylbenzapine derivatives in
Table 1 are similar in both the D
1
R and D R structures or a little better
5
in D R as a consequence of the slightly stronger hydrogen bonding and
5
π-π hydrophobic interactions. In this context, the docking scores do not
align with the selectivity trends derived from the experimental binding
affinities. This is at least partially attributable to the modest nature of
the observed D
1
R/D R experimental selectivities of the ring C haloge-
5
nated analogs. Furthermore, D
1
R and D R are very similar structurally
5
in the ligand binding pocket, which provides justification for the close
computational binding energies for most of the compounds in these two
target sites. The docking outcomes for the compounds in Table 1 in-
volved the R enantiomers whereas the affinity data were obtained with
racemic mixtures and this could also have an impact on the match
between the experimental and computational results. Docking simula-
tions were investigated with the S enantiomers, however, they gener-
ated similar trends compared to the R enantiomers with, in general,
slightly worse predicted binding energies in both the D
targets.
1
R and D R
5
In conclusion, this study extends the available SAR information on
1
-phenylbenzazepines as D
ring C ortho halogen substituents. As is evident from examination of the
data, the compounds in this study maintain selectivity for D -like re-
ceptors over D R, with modest selectivity for either D R or D R. As
compared to known 1-phenylbenzapeine D R-like tools such as 1 and 2,
it is apparent that the ortho halogen group does not significantly im-
prove D R or D R affinity. However, one of the findings from this work
is that compounds with di-ortho-halo substituents (i.e. C2′/C6′ sub-
stitution) favor binding to D R, whereas compounds with a mono-ortho-
halo (C2′) substituent favor D R binding over D R. In addition, the SAR
data suggests that the most favorable outcome for good D R affinity is
1
R-like ligands with regards to the effect of
1
2
1
5
1
1
5
5
1
5
1
to have either mono-ortho-halogenation in tandem with N-alkyl sub-
stitution or di-ortho-halogenation without N-alkyl substitution.
Evaluation of the functional activity of 10a reaffirms the idea that
an 8-hydroxy-7-methoxy moiety favors antagonist rather than agonist
activity. This result is in line with the generally accepted view that a
catechol motif is required for agonist activity in the 1-phenylbenzaze-
pine scaffold.
Fig. 2. Docked poses of compounds 7 (blue carbon atoms), 10a (green carbon
atoms) and 10c (pink carbon atoms) in A - the D
1
R target and B - the D R target.
5
The receptor targets are depicted by secondary structure elements and grey
carbon atoms for select residues. Key quaternary N – Asp salt bridges are de-
picted by the pink dashed lines, H-bonding interactions by the yellow dashed
lines, aromatic H-bonding by the turquoise dashed lines and π-π stacking by the
blue dashed lines. Docking studies were performed with the R enantiomers.
Our molecular docking studies revealed interactions that were re-
levant for affinity of the molecules at D
1
R and D R, but were unable to
5
resolve interactions necessary for the observed modest D R sub-type
1
selectivity of 7, 10a and 10c. Examination of larger sets of compounds
with ortho halogenated patterns in future, including enantiopure ana-
logs, may provide a larger body of data to aid in the challenging opti-
Schrödinger Glide methodology in Standard Precision (SP) mode. Using
this approach, the Glidescore scoring function provided an estimate of
the ligand binding affinities for the highest ranked poses of the ligand
mization of these ligands towards D
selectivity.
1
R or D R potency and sub-type
5
series in the D
1
R and D R targets. The binding poses for the compounds
5
7
, 10a and 10c (docked as the R enantiomers), which gave the best D R
1
experimental affinities, are depicted in Fig. 2A and 2B.
Declaration of Competing Interest
Compounds 7, 10a and 10c give very similar docked poses in the
D
1
R binding pocket as shown in Fig. 2A with binding energies in the
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influ-
ence the work reported in this paper.
range −7.8 kcal/mol to −8.2 kcal/mol. The docked poses display the
quaternary N - Asp103 salt bridge, H-bonding interactions of the ligand
hydroxyl group to the Asn292 side chain, and for compound 7 with an
additional hydroxyl group in the catechol moiety there is also a H-bond
to the Ser198 sidechain, as well as an aromatic H-bond involving the
pendant phenyl group and Ser188.
Acknowledgments
In the D
5
R binding site, the main receptor-ligand interactions for the
This publication was made possible by Grant Number
1SC1DA049961-01 from the National Institutes of Health (NIH). Its
contents are solely the responsibility of the authors and do not ne-
docked poses of compounds 7, 10a and 10c comprise the quaternary N
-
Asp120 salt bridge, hydrogen bonding interactions of the ligand hy-
droxyl group to the Asn316 or Ser229 sidechain, and again for com-
pound 7 there is another hydrogen bond with its second catechol hy-
droxyl group to Ser233, as well as π-π hydrophobic interactions
involving the ligand aromatic rings with Phe312 and Trp116.
cessarily represent the official views of the NIH or its divisions. K de-
i
terminations, and receptor binding profiles were generously provided
by the National Institute of Mental Health's Psychoactive Drug
Screening Program, Contract
# HHSN-271-2008-00025-C (NIMH
4