16
N. Adhikari et al. / Bioorg. Med. Chem. xxx (2016) xxx–xxx
higher experimental activity of A13. Interactions with Tyr74, Le82,
Tyr142 and Phe148 are observed for A13 but these are either
absent or less frequent in A21. Interestingly, A21 showed more
interactions with residues like His85, Ala86, Phe87, Ala88,
Leu116 and Val117. To elaborate these interactions, the timeline
representations of protein–ligand contacts of these two ligands
are observed (Fig. S8, Supporting information). The plot reveals
that these interactions of A21 are not only more frequent than
A13 but also complementary to each other (i.e., in 10 ns MD run,
the ligand is found to interact with some of these residues). Inter-
estingly, these amino acid residues are part of S10 pocket of MMP-2
and it has earlier been reported that more number of interactions
with the hydrophobic pocket may ensure the higher selectivity
towards MMP-2 against MMP-8.38 The docking interaction of A21
suggested that the 3,5-bis-trifluromethylbenzyl residue interacts
with His85, Ala88 and Phe87. It may, therefore, be hypothesized
that these interactions are responsible for non-selectivity of A21
towards MMP-8. Two of the non-conservative residues between
MMP-2 and MMP-8 (Thr143 of MMP-2 or Ala220 of MMP-8 and
Thr145 of MMP-2 or Arg228 of MMP-8) have also been recognized
for the differences in ligand affinities between MMP-2 and MMP-8.
for immunofluorescence as well as migration and invasion assays
for the active compounds.
3.10. Immunofluorescence assay for cellular MMP-2 expression
To explore the potential of these synthesized molecules to inhi-
bit MMPs in human cancer cell lines, immunofluorescence micro-
scopy technique was adopted to find out cellular localizations of
MMP-2 enzyme in the treated and untreated cell lines. The A549
cell line (1 ꢀ 106 cells) was cultured in chamber slides and treated
with a fixed dose (50 lM) of three compounds (Compd A9, A13 and
A21) for 24 hours. The best active A13 is chosen along with A9
since the later is the most selective MMP-2 inhibitor with the ali-
phatic side chain. The A21, on the other hand, is the most selective
MMP-2 inhibitor of this series. After 24 h, cells were washed with
phosphate buffer saline (PBS) and fixed in chilled methanol and
permeabilized in 0.01% triton X 100. Nonspecific bindings were
blocked by bovine serum albumin (BSA) in phosphate buffer saline
(PBS), followed by incubation with MMP-2 antibodies. Expression
was detected with conjugated secondary antibody. For nucleus
counterstaining, 40-6-diamidino-2-phenylindole (DAPI) was used.
The expressions of MMP-2 antibodies in untreated control and
treated cells are presented in Figure 8A.
It is evident from Figure 8A that all treated cells showed the
lower expression of MMP-2 comparing with the untreated cells.
The mean fluorescence reading was recorded. The total intensity
of fluorescence was measured by calculating the corrected total
cell fluorescence (CTCF) of the control and samples are provided
in Supplementary materials (Fig. S11). The cellular MMP-2 expres-
sion of A13 treated cells is slightly lower than that of A9. The over-
all MMP-2 expression is reduced upto 60.50% in A13 treated cells
whereas the intensity is lowered upto 52.10% in A9 treated cells.
Notwithstanding the lowest observed binding affinity among these
three derivatives, the most prominent inhibition is obtained for
A21. This compound reduces MMP-2 expression upto 78.90% com-
paring with that of the control. The DAPI and phase contrast dia-
grams (Fig. 8A) depict that the designed compounds did not
produce considerable DNA fragmentations and morphological fea-
When the docking interactions predicted a possible p-donor inter-
action with Thr143 in both these ligands, no information of these
interactions is deduced from the MD simulation. On the other
hand, interaction with Thr145 residue is unlikely since the ligand
poses are located away from this residue. The analyses of the
ligand root mean square fluctuations (RMSFs) (Fig. S9, Supplemen-
tary material) further reveal that the benzyl residues of both
ligands have the higher fluctuations (or variable interactions)
whereas the rest of these molecules have the lower fluctuations
(or constant interactions). However, the average RMSF of 3,5-bis
trifluromethylbenzyl moiety of A21 is slightly less than the benzyl
moiety of A13. It may, therefore, be interpreted that the better sta-
bility in the terminal residues may be a determining factor for both
the affinity and the selectivity of this series of molecules.
3.9. Cytotoxicity and flow cytometry apoptotic assays
The designed compounds having MMP-2 IC50 values < 105 nM
were tested for cytotoxicity against A549 cell line by 3-(4,5-
dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT)
assay. The cytotoxicity results are shown in the Supporting infor-
mation (Table S7). All these compounds showed lower cytotoxicity
against A549 cell line and most of these compounds showed cell
ture changes at 50 lM concentrations.
3.11. Wound healing migration assay
To investigate anti-migratory properties of the designed com-
pounds, migration assay was performed with A549 cell line which
viability 80–90% at concentration of 100
these inhibitors exhibited 50% or more cytotoxicity up to 300
(Table S7, Supporting information). This compound showed an
IC50 value of 270 M. The results of A13 and A21 treated A549 cells
at four different concentrations between 0 and 300 M are pro-
vided in the Supporting information (Fig. S10). None of these two
compounds significantly reduce the cell viability compared with
the untreated control. It was earlier reported that MMPIs may
not exhibit cytotoxicity even at the higher concentrations.89,90 To
understand the mechanistic involvement of apoptosis in the anti-
proliferative activity, A13 and A21 were tested in annexin-V/pro-
pidium iodide (PI) flow cytometry assay (Fig. S10, Supporting infor-
l
M. Except A21, none of
was treated with 50 lM of compounds A9, A13 and A21 for 48 h
lM
after a wound was made on the monolayer of cells. The observa-
tions are illustrated in Figure 8B. It demonstrates that the control
cells extensively migrated into the denuded area. However, the cell
migration was significantly reduced by the designed compounds.
The inhibition was most prominent for the compounds A13 and
A21 whereas A9 also reduced the migration but overall effect
was lower than that of these two compounds.
l
l
3.12. Invasion assay
Inhibitions of A549 cell invasion by the designed compounds
were measured using fluorimetric QCM ECMatrix cell invasion
assay (ECM 555, Millipore). Some higher active designed com-
pounds (A6, A9, A11, A13, A21 and A23) were investigated for their
anti-invasive properties. The total numbers of migrated cells are
graphically represented in Figure 8C. It was observed that all of
these compounds were able to reduce the total number of
migrated cells. The maximum inhibition was observed for the com-
pound A13 (55.8 2.7%), followed by the compounds A6
(52.0 3.0%) and A21 (51.0 1.5%). The other three compounds
showed less than 50% anti-invasive properties (A9: 37.0 2.6%,
mation). Two different doses of 100 and 200
lM were chosen for
this assay. At 100 M concentration, both inhibitors showed min-
l
imal apoptotic as well as necrotic effects. However, the necrotic
effects predominate in both compounds. At the higher concentra-
tions (200 lM), necrotic effect increases approximately 2.5%
whereas apoptotic effects remain unchanged. Two conclusions
may be drawn from these studies—(I) these compounds are not
cytotoxic below 100
200 M may elicit undesirable necrotic effects in the cell lines.
Therefore, a fixed noncytotoxic concentration of 50 M was chosen
lM and (II) concentrations higher that
l
l