From 1983 to 1993, Diabetes Control and Complications
Trials (DCCT) were conducted that established the role of
hyperglycemia in the pathogenesis of chronic diabetic
complications [1]. Hyperglycemia is reported to show its
complications via various mechanisms, including increased
aldose reductase activity, non-enzymatic glycation and
glycoxidation, activation of protein kinase C (PKC) and
oxidative nitrosative stress. Moreover, reactive oxidative and
nitrogen species induce activation of mitogen-activated protein
kinases (MAPKs), poly (ADP-ribose) polymerase (PARP) and
other downstream targets [2]. The complex etiopathology of
diabetic complications and unifying mechanism discussed by
Brownlee et al exhibit that the use of multi-targeted agents with
multiple complementary biochemical activities would be a
promising therapeutic option for the intervention of the disease
over single targeted drugs [2,3]. Moreover, the encouraging
results provided by a clinical candidate ‘benfotiamine’ (Figure 1)
shows that it blocks three pathways, i.e. hexosamine pathway,
AGEs pathway and protein kinase C pathway [4].
Based on the above hypothesis, the present study was
undertaken to design and evaluate novel dual inhibitors for ALR2
and PARP-1 using structure based design approach. The use of
molecular docking has often been unfruitful due to the flaws in
the protein-ligand binding free energy function to score putative
inhibitors leading to poor correlation between computational and
experimental measurements of biological potencies. To cope with
this problem, molecular dynamics simulation was incorporated
into the structure based design protocol. The designed inhibitors
were further synthesized and biologically evaluated using in-vitro
enzyme assay against ALR2 and PARP-1.
Prior to docking and molecular dynamics simulations, the
crystal structures of ALR2 and PARP-1 proteins complexed with
respective inhibitors were thoroughly studied for their interaction
pattern within the active sites. The inspection of interaction
pattern of inhibitors within ALR2 active site revealed that the
hydrogen bond interaction with His110 and Tyr48 amino acid
residues remain conserved (Table S1, Supplementary Material).
The formation of π-π interactions with Trp111 is highly
conserved and seems to play important role in binding. Hydrogen
bond interaction with Leu300 and π-π interaction with Trp20
were also present in some of the crystal structures. Moreover,
studies suggest that interaction with amino acid Leu300 is
important as it is part of the selectivity pocket of the ALR2
binding site. In PARP-1 crystal complexes, it was observed that
hydrogen bond interactions with Gly202 and Ser243 amino acid
residues are conserved in the available crystal structures (Table
S2, Supplementary Material). Additionally, π-π interactions with
aromatic residues Tyr246 and His201, and π-cation interaction
with Lys242 amino acid residue were retained in most of the
crystal complexes.
Diabetic retinopathy is a common reason of vision loss
characterized by retinal capillary cell loss, capillary basement
membrane thickening, increased vascular permeability and
increased leukocyte adhesion to endothelial cells. The selective
destruction of retinal pericytes has been linked to accumulation
of sorbitol which results in osmotic stress due to excessive
hydration, gain of Na+ and loss of K+ ions [5, 6]. In addition,
literature reports highlight that increased aldose reductase activity
is responsible for enhancing oxidative stress, up-regulates retinal
vascular endothelial growth factor (VEGF) and activation of
PARP in diabetic retinal cells which may lead to cataract
formation and diabetic retinopathy [7-10]. Moreover, the two
enzymes: ALR2 and PARP-1 were found to co-express in the
retina cells [5,11,12]. The complexity of the disease as well as
the above reports suggest that dual inhibition of both the enzymes
would provide an efficient strategy to ameliorate the pathology of
diabetic retinopathy.
From the information extracted from ALR2 active site, it was
observed that the presence of an acidic moiety in inhibitor was
required for interaction with Tyr48, His110 and Trp111 amino
acid residues. For PARP-1 it was observed that an acceptor and
donor feature is required for interaction with Gly202 and Ser243
amino acids. Moreover, the pharmacophores in the clinical trial
molecules as well as marketed drugs against ALR2 suggest that
acidic group (for interaction with basic amino acid His110) and
large hydrophobic/aromatic group are required for interaction. In
PARP inhibitors, paired acceptor donor is present in the form of
amide linkage within the molecule. Therefore, it was thought that
the presence of thiazolidine-2,4-dione (TZD) ring in the inhibitor
would serve as acidic moiety for ALR2 and would also provide
acceptor and donor paired feature for PARP-1. In addition, it has
been reported that a hydrophobic group in inhibitor is required to
occupy the specificity pocket of the ALR2 active site lined by
Trp111, Phe122 and Leu300 amino acid residues [22]. For
PARP-1 inhibitor, hydrophobic group is required for interaction
with aromatic residues Tyr246 and His201. To sustain
hydrophobic interaction between the designed molecules and
active sites of both enzymes, substitution of benzyl along with
hydrophobic substituents was considered to be beneficial.
Keeping the above facts and synthetic feasibility in mind, a series
of novel compounds (5a-5l) containing TZD ring and substituted
benzyl groups linked via indole ring were designed. These
designed molecules were duly synthesized and optimized to
obtain potent dual ALR2 and PARP-1 inhibitors (Figure 2).
Aldose reductase (ALR2) is a cytosolic NADPH-dependent
oxidoreductase acting as first and rate controlling enzyme of
polyol pathway [13]. Physiologically, it catalyzes the reduction
of various aldehydes and carbonyls, primarily glucose to sorbitol.
In diabetes, the excess intracellular sorbitol produced by the
over-activation of ALR2 leads to osmotic damage to cells that
eventually leads to diabetic complications [14]. Numerous ALR2
inhibitors have been developed to prevent retinal and neuronal
damage, including tolrestat, epalrestat, fidarestat and ranirestat,
etc (Figure 1). Out of these, epalrestat has been approved in
Japan while tolrestat was withdrawn from market owing to
hepatotoxicity. Fidarestat and ranirestat are under clinical
evaluation [15]. Dual inhibitors of ALR2 and antioxidants/
protein tyrosine phosphatase (PTP1B) have also been reported in
literature [16,17].
Poly (ADP-ribose) polymerase (PARP-1) is a nuclear enzyme
that regulates DNA repair, cellular division, differentiation, DNA
replication, transformation, gene expression and amplification,
mitochondrial function, and cell death [18]. The overexpression
of PARP in the retina of diabetic rats has been reported to occur
due to DNA damage induced by cell death [19]. Though none of
the inhibitor has been reported with activity in diabetic
complications, olaparib is one PARP inhibitor, which is in reach
the market implicated in advanced ovarian cancer. Others are
under clinical trials, including veliparib and rucaparib (Figure 1)
[20]. Dual inhibitors of PARP-1 and dihydroorotate
dehydrogenase (DHODH) have been reported in literature with
benzimidazole nucleus [21].
The designed inhibitors were subsequently subjected to
docking analysis in ALR2 (PDB ID: 1US0) and PARP-1 (PDB
ID: 2RD6) proteins. As depicted in Figure 3A, the docked pose
of designed inhibitor in ALR2 active site revealed that the
carbonyl group of the TZD moiety formed hydrogen bonding
interaction with Tyr48 and His110, while the indole ring forms π-
Chadha, Navriti
