A library of hybrid molecules was procured by the combination of triazine-indole adduct with morpholine/piperidine/pyrrolidine and pyrazole/pyrimidine/oxindole moieties. Enzyme immunoassays on cyclooxygenase-1 (COX-1) and cyclooxygenase-2 (COX-2) identified compound 6 having an IC50 value of 20 nM for COX-2 and 3000 nM for COX-1. The significant reduction in the formation of prostaglandin E2 in the lipopolysaccharide-treated (COX-2-activated) human whole blood, almost no change in the production of thromboxane B2 in the calcium ionophore-treated (COX-1-activated) sample of human whole blood, and the mechanistic studies on Swiss albino mice ensured that compound 6 is selective for COX-2. The association constant (Ka) of compound 6 with COX-2 was found to be of the order of 0.48 × 106 M-1. The diffusion spectroscopy experiments and relaxation time (T1) calculations of compound 6 in the presence of COX-2 assisted in identifying the site-specific interactions of 6 with the enzyme, and these results fall into nice correlation with the theoretical data obtained from molecular docking and quantitative structure-activity relationship studies. With maximum tolerable dose >2000 mg kg-1, compound 6 made 68 and 32% reduction in formalin-induced analgesia and carrageenan-induced inflammation in Swiss albino mice.
A library of hybrid molecules was procured by the combination of triazine-n class="Chemical">indole adduct with morpholine/piperidine/pyrrolidine and pyrazole/pyrimidine/oxindole moieties. Enzyme immunoassays on cyclooxygenase-1 (COX-1) and cyclooxygenase-2 (COX-2) identified compound 6 having an IC50 value of 20 nM for COX-2 and 3000 nM for COX-1. The significant reduction in the formation of prostaglandin E2 in the lipopolysaccharide-treated (COX-2-activated) human whole blood, almost no change in the production of thromboxane B2 in the calcium ionophore-treated (COX-1-activated) sample of human whole blood, and the mechanistic studies on Swiss albino mice ensured that compound 6 is selective for COX-2. The association constant (Ka) of compound 6 with COX-2 was found to be of the order of 0.48 × 106 M-1. The diffusion spectroscopy experiments and relaxation time (T1) calculations of compound 6 in the presence of COX-2 assisted in identifying the site-specific interactions of 6 with the enzyme, and these results fall into nice correlation with the theoretical data obtained from molecular docking and quantitative structure-activity relationship studies. With maximum tolerable dose >2000 mg kg-1, compound 6 made 68 and 32% reduction in formalin-induced analgesia and carrageenan-induced inflammation in Swiss albino mice.
Inflammation
is a preeminent host defense to injury, infectious
agents, and autoimmune responses.[1] However,
the chronic n class="Disease">inflammation leads to the emergence of various diseases
such as rheumatoid arthritis, neurodegenerative disorder, and cancer
and cardiovascular diseases.[2] The most
important markers of inflammation include cytokine receptors, nitric
oxide synthase (NOS), nuclear factor kappa-B, chemokines, tumor necrosis
factor alpha, interferons, and proinflammatory enzymes cyclooxygenase-2
(COX-2) and lipoxygenase.[3] Among these
mediators, cyclooxygenases or prostaglandin endoperoxide synthases,
taking part in arachidonic acid (AA) metabolism for the formation
of inflammation causing prostaglandins,[4] are the decisive players. Of the two isoforms of cyclooxygenases,
cyclooxygenase-1 (COX-1) and COX-2 are almost identical in structure
but have distinct biological functions.[5] The inducible enzyme COX-2 mediates the synthesis of proinflammatory
prostaglandins and thromboxanes,[6] and it
becomes the major therapeutic target for the treatment of inflammatory
diseases. On the other hand, housekeeping enzyme COX-1 helps in homeostasis
and blood clotting. As far as the treatment of inflammatory diseases
is concerned, it has mostly been dependent on the use of nonsteroidal
anti-inflammatory drugs including COX-1/2 nonselective inhibitors
such as aspirin, ibuprofen, indomethacin, and diclofenac and selective
COX-2 inhibitors (COXIBS) such as celecoxib.[7] However, the severe side effects including gastrointestinal lesions,
renal injury, and cardiovascular diseases associated with the clinical
usage of these therapeutic agents[8] have
necessitated the search for new chemical entities with higher efficacy
and low/no side effects.
The pattern of exhibiting diverse biological
activities by the
heterocyclic moieties has imparted them paramount importance in constituting
the skeleton of several medicinally significant compounds.[9] Specifically, the aza-heterocycles such as indole,
n class="Chemical">pyrimidine, pyrazole, morpholine, piperidine, pyrrolidine, and triazine
are serving as templates of a number of clinically used anti-inflammatory,
antifungal, antileukemic, and neuroprotective agents[10] and antiamoebic, anticancer, antileishmanial, antimalarial,
antiviral, antitubercular, carbonic anhydrase inhibitor, cathepsin
B inhibitor, and antimicrobial agents.[11−14] In the context of these reports,
it was envisaged that a hybrid molecule obtained by the combination
of two/three aza-heterocycles may prove more efficacious than those
drugs, which are made up of the individual aza-heterocycle moieties.
Therefore, in continuation to the efforts for the development of indole–pyrazole/indolinone–chrysin-based
conjugates as selective COX-2 inhibitors (A–D Figure ),[15] it was further planned to replace the chrysin moiety of the molecules
with the substituted triazine. Hence, the replacement of the red colored
part of molecules A–D with the blue colored component (consisting
of indole, triazine, morpholine, piperidine, pyrrolidine, pyrimidine,
and pyrazole heterocycles) led to the design of compounds 2–22 (Figure ). These
compounds were synthesized and screened for anti-inflammatory activity
by using enzyme immunoassays and animal models. COX-2 as the cellular
target of the molecules was confirmed by the in vivo mechanistic experiments
supported by the physicochemical and the molecular modeling studies.
Figure 1
Previously
reported molecules A–D and the library of newly
designed compounds 2–22.
Previously
reported molecules A–D and the library of newly
designed compounds 2–22.
Results and Discussion
Chemistry
The reaction of morpholine
and n class="Chemical">cyanuric chloride provided di- and monosubstituted products 1a and 1b (Scheme ). Taken in acetonitrile (ACN) in the presence of NaH,
compound 1a was treated with indole-3-carboxaldehyde
at room temperature (rt) and compound 2 was procured.
Treatment of compound 2 with 1-(2,6-dichlorophenyl)-1,3-dihydroindol-2-one(indolinone)/1,3-dihydroindol-2-one(oxindole)/N,N-dimethylbarbituric acid/1-(3-chlorophenyl)-3-methyl-2-pyrazolin-5-one
under microwave irradiation in the presence of piperidine afforded
compounds 3–6 (Scheme ).
Scheme 1
Synthesis of Compounds 2–6
Reagents and conditions: (a)
Et3N, acetone, 4 h, 0 °C to rt; (b) NaH, ACN, 3 h,
rt; and (c) MWI, CHCl3, piperidine, 2 h, 100 °C.
Synthesis of Compounds 2–6
Reagents and conditions: (a)
Et3N, n class="Chemical">acetone, 4 h, 0 °C to rt; (b) NaH, ACN, 3 h,
rt; and (c) MWI, CHCl3, piperidine, 2 h, 100 °C.
Similar to the reaction of 1a, treatment
of compound 1b with indole-3-carboxaldehyde provided
compound 7. Reaction of compound 7 with
n class="Chemical">1-(3-chlorophenyl)-3-methyl-2-pyrazolin-5-one/N,N-dimethyl barbituric acid/oxindole/indolinone
under microwave irradiation in the presence of piperidine afforded
compounds 8–12, respectively. Likewise, compounds 13–17 were prepared when compound 7 was
reacted with 1-(3-chlorophenyl)-3-methyl-2-pyrazolin-5-one/N,N-dimethyl barbituric acid/oxindole/indolinone
in the presence of pyrrolidine under microwave irradiation (Scheme ). Interestingly,
the incorporation of a heterocycle moiety at the CHO group of 7 and replacement of its Cl with piperidine/pyrrolidine occurred
in one pot. However, stepwise reaction of 7, first with
piperidine/pyrrolidine and then with the heterocycle moiety or vice
versa, was also successful.
Scheme 2
Synthesis of Compounds 7–17
Reagents and conditions: (a)
NaH, ACN, 3 h, rt and (b) microwave irradiation (MWI), CHCl3, piperidine/pyrrolidine, 2 h, 100 °C.
Synthesis of Compounds 7–17
Reagents and conditions: (a)
NaH, n class="Chemical">ACN, 3 h, rt and (b) microwave irradiation (MWI), CHCl3, piperidine/pyrrolidine, 2 h, 100 °C.
Compounds 18–20 were obtained by heating compound 7 (1 mmol) with 1-(3-chlorophenyl)-3-methyl-2-pyrazolin-5-one/N,n class="Chemical">N-dimethyl barbituric acid/oxindole (1
mmol) at 155–160 °C for 25–30 min (Scheme ).
Scheme 3
Synthesis of Compounds 18–20
Reaction of piperidine (2 mmol) with n class="Chemical">cyanuric chloride
(1 mmol)
provided product 23 (Scheme ). Treatment of compound 23 with
indole-3-carboxaldehyde resulted into the formation of compound 21. Compound 21 on further reaction with 1-(3-chlorophenyl)-3-methyl-2-pyrazolin-5-one
under microwave irradiation in the presence of piperidine provided
compound 22 (Scheme ).
Scheme 4
Synthesis of Compounds 21–22
Interestingly, the treatment
of cyanuric chloride with n class="Chemical">indole-3-carboxaldehyde
provided compound 24 (Scheme ). However, because of the high melting point
(>300 °C) and insolubility of compound 24 in
most
of the polar and nonpolar solvents, we were not able to make further
derivatization of this compound.
Scheme 5
Synthesis of Compound 24
Reagents and conditions: (i)
NaH, ACN, 0 °C, 0.5 h.
Synthesis of Compound 24
Reagents and conditions: (i)
NaH, n class="Chemical">ACN, 0 °C, 0.5 h.
All the compounds
were characterized by using NMR, IR, and high-resolution
mass spectral techniques. NMR spectral data have unambiguously assigned
Z-configuration at the double bond across C9–C10 (Figure ). On the basis of
various n class="Chemical">1D and 2D NMR experiments including 1H, 13C, nuclear Overhauser enhancement spectroscopy (NOESY), correlation
spectroscopy, heteronuclear single quantum coherence (HSQC), and heteronuclear
multiple bond correlation (HMBC) (Figures and 3), all the proton
and carbon resonance frequencies of compound 6 were assigned.
Characteristically, the peak at δ 8.12 is assigned to olefinic
proton (H-9). The NOESY spectrum of compound 6 clearly
showed the presence of NOE between H-9 and H-15 as well as between
H-9 and H-1 (Figures and S33), which confirmed the Z-configuration
at the bridged C=C (C9–C10). Likewise, the H and C resonances
of other compounds and their geometry across the bridged C=C
bond were assigned.
Figure 2
Overlay of HSQC (red contours) and HMBC (blue contours)
NMR spectrum
of compound 6.
Figure 3
1H–1H NOESY NMR spectrum of compound 6. Inset: expansion of a part of the spectrum.
Overlay of HSQC (red contours) and HMBC (blue contours)
NMR spectrum
of compound 6.1H–n class="Chemical">1H NOESY NMR spectrum of compound 6. Inset: expansion of a part of the spectrum.
Biological Studies
In Vitro COX inhibitory Activity
By using the protocol
available with the assay kits,[16] COX-1
and n class="Gene">COX-2 inhibitory activities of compounds 2–22 were checked. The enzyme inhibition assay was
based on the quantification of the prostaglandins produced by the
COX in AA metabolism, in the absence and presence of the test compounds.
All the compounds were tested in triplicate at 10–4, 10–5, 10–6, 10–7, and 10–8 M concentration, and the 50% inhibitory
concentration (IC50) for each of the compounds was calculated.
The difference in the IC50 of the three enzyme assays was
<3%, and the average of the three is reported in Table .
Table 1
IC50 (μM) of Compounds 2–22 against
COX-1 and COX-2
IC50 (COX-1)/IC50 (COX-2).
IC50 (COX-1)/IC50 (n class="Gene">COX-2).
Appreciable inhibition of the COX-2 activity was observed
in the
presence of compound 2, and its IC50 was calculated
0.03 μM. Desirably, it was observed that compound 2 is selective for n class="Gene">COX-2 over COX-1 and that the selectivity index
was higher than that seen in the case of diclofenac and less than
that of celecoxib. The presence of two morpholine moieties in compound 2 seems enviable for COX-2 inhibition because the analogue
compounds 7, 8, 13, and 21 exhibit higher IC50 than that of compound 2. Compounds 2, 7, 8, 13, and 21 were further derivatized by
incorporating a heterocycle moiety at their CHO group.
Compounds 3–6 (obtained by the derivatization
of compound 2) exhibited wide spectrum of their COX-2
inhibitory profile. Compound 6 has an IC50 of 0.02 μM for n class="Gene">COX-2, which is 2-fold higher than that of
celecoxib and is comparable to the IC50 of diclofenac.
The selectivity index 150 of compound 6 for COX-2 was
much better than that of diclofenac and indomethacin. Compound 3 with an IC50 of 0.4 μM was also found to
exhibit appreciable inhibition of COX-2, whereas compound 4 exhibited an IC50 of 1 μM for COX-2. The poor solubility
of compound 5 in the assay medium did not allow its screening
in the enzyme immunoassays.
Modification of compound 8 to compounds 9–12 also resulted into a significant
variation in their IC50 for COX-2. Compounds 10, 11, and 12 exhibited IC50 values
of 0.08, 0.6, and 0.05
μM, respectively, for n class="Gene">COX-2. The selectivity index > 200
for
compound 12 was appreciably higher than for diclofenac
and indomethacin. However, the incorporation of a heterocycle moiety
at the CHO group of compound 13 resulted into relatively
less potent compounds 13–17. The derivatization
of compound 7 into compounds 18–20 improved their COX-2 inhibitory profile. Conversion of compound 21 to compound 22 made 70-fold increases in its
IC50 for COX-2 (Table ).
Apparently, the compounds 6, 18, 12, 17, and 22 obtained
by the incorporation
of n class="Chemical">1-(3-chlorophenyl)-3-methyl-2-pyrazolin-5-one on compounds 2, 7, 8, 13, and 21, respectively, resulted into better inhibition of COX-2.
Moreover, in addition to the contribution of the pyrazole moiety,
the presence of two morpholine units in compound 6 also
seems responsible for the higher efficacy of this compound. The role
of a morpholine moiety was also apparent from the comparison of IC50 values of compound 3 with that of compounds 9 and 14. However, for compounds 4, 10, 15, and 20, compound 10 with morpholine and piperidine moieties along with oxindole
exhibited better IC50 (0.08 μM) for COX-2. In this
group, compound 4 with two morpholine units along with
oxindole has an IC50 value of 1 μM. Conspicuously,
the presence of a pyrrolidine moiety in compounds 14, 15, and 17 increased the IC50 in comparison
to that of compounds 3, 9, 4, 10, 6, and 12. Therefore,
the analysis of the data given in Table indicated that the morpholine/piperidine/pyrrolidine
moiety contributes for the COX-2 inhibition in association with the
heterocycle unit present on C-3 of indole. Hence, the prevalence of
significant structure–activity relationship in these compounds
may not put them into the category of PAINS assay.[17] Overall, the presence of two morpholine units and a pyrazole-bearing
indole moiety on the triazine template was optimized for the COX-2
inhibitory activity. On the basis of the preliminary investigations,
compounds 2, 3, 4, 6, 10, 12, 17, 18, and 22 were further screened over the animal models
for their analgesic and anti-inflammatory activity.
Human Whole Blood COX-1 and COX-2 Assay
Calcium ionophore
(n class="Chemical">A23187)-stimulated production of thromboxane
B2 (TXB2) in whole blood platelets was used
to measure the activity of COX-1, whereas prostaglandin E2 (PGE2) production in lipopolysaccharide (LPS)-stimulated
whole blood was used to assess the COX-2 activity.[18] In both the cases, enzyme-linked immunosorbent assays were
performed.[16,19] These assays measured the amount
of TXB2 and PGE2 in the serum that was produced
in the presence and absence of the compound.
A23187-stimulation of n class="Species">human
whole blood
resulted in the increase in TXB2 production compared with
the control blood sample (Table ). Addition of 1 μM of compounds 6, 10, 12, and 17 to the blood
sample did not affect ionophore-stimulated TXB2 production,
indicating that compounds 6, 10, 12, and 17 exhibited almost negligible inhibition of COX-1.
On the other hand, LPS stimulation of human whole blood increased
the PGE2 production compared with the control blood sample
(Table , Figure ), and the addition
of 1 μM of compounds 6, 10, 12, and 17 significantly decreased the PGE2 production, hence indicating that compounds 6, 10, 12, and 17 inhibited
COX-2.
Table 2
Results of Enzyme Immunoassays Showing
the Role of Compounds 6, 10, 12, and 17 (Final Concn 1 μM) in Inhibition of Calcium
Ionophore-Stimulated TXB2 and LPS-Stimulated PGE2 in Whole Blood
TXB2 (ng/mL)
PGE2 (ng/mL)
–calcium
ionophore
+calcium ionophore
–LPS
+LPS
control
0.16
2.00
0.145
1.98
indomethacin
1.05
1.35
6
1.85
0.91
10
1.55
1.02
12
1.75
0.95
17
1.70
1.00
Figure 4
Graphical representation of the data given in Table . (A) TXB2 production
in the presence of compounds is the same as in the control experiment,
indicating that compounds did not inhibit COX-1. (B) PGE2 production in the presence of compounds is significantly lower than
the control experiment indicating the inhibition of COX-2.
Graphical representation of the data given in Table . (A) TXB2 production
in the presence of compounds is the same as in the control experiment,
indicating that compounds did not inhibit n class="Gene">COX-1. (B) PGE2 production in the presence of compounds is significantly lower than
the control experiment indicating the inhibition of COX-2.
In Vivo Biological Studies
In vivo
biological experiments were performed with the Swiss albino mice of
either sex weighing 25–30 g. The study design was duly approved
by the Institutional Animal Ethical Committee (IAEC). The n class="Chemical">formalin-induced
hyperalgesia and carrageenan-induced paw inflammation models were
used to study the analgesic and anti-inflammatory activities, respectively,
of the compounds 2, 3, 4, 6, 10, 12, 17, 18, and 22.[20] For
analgesic activity, 11 groups of animals comprising five animals in
each group were used. Group 1 was administered vehicle, group 2 diclofenac
(10 mg kg–1), and groups 3–11 received compounds 2, 3, 4, 6, 10, 12, 17, 18, and 22, respectively, at the dose of 10 mg kg–1. All
the compounds were administered intraperitoneally (i.p.) 30 min before
formalin injection. For studying if COX, LOX, and nitric oxide pathways
are the potential targets of the compounds, three groups including
five animals per group were used. These groups received substance
P (COX and LOX pathway stimulator), l-arginine (NO donor),
and l-NAME (NOS inhibitor) 30 min before the most potent
compound 6 was given and then were injected formalin
after 30 min. To check if voltage-gated sodium channels and calcium
influx processes are targeted by the compounds, the animals were pretreated
with veratrine and A23187, respectively. For anti-inflammatory activity,
three groups were used. The first group served as control, the second
group was treated with standard drug indomethacin (10 mg kg–1), and the third received the most potent compound 6, followed by carrageenan.
Formalin-Induced Analgesia
Administration
of n class="Chemical">diclofenac and compounds 2, 3, 4, 6, 10, 12, 17, 18, and 22 significantly decreased
the number of flinching in the inflammatory phase as compared to the
control group (Figure , Table ). The effect
of compounds 6, 12, and 17 was
relatively higher in the series, although the difference in the analgesic
effect of the various other analogues was statistically nonsignificant.
Compounds 6, 12, and 17 were
found to decrease formalin-induced analgesia by 68.24, 74.25, and
69%, respectively. The analgesic effect of compound 6 was significantly attenuated on pretreatment with substance P, whereas
pretreatment with nitric oxidedonor, l-arginine, or NOS
inhibitor l-NAME did not alter the analgesic effect of compound 6 (Figure ). Because substance P is known to stimulate COX-2 and LOX pathways[15b] and compound 6 did not inhibit
5-LOX activity (as determined by the enzyme immunoassay, IC50 > 100 μM), the analgesic effect of this compound was probably
due to the inhibition of the COX-2 pathway. Because the analgesic
effect of compound 6 was not altered on l-arginine
or l-NAME pretreatment, the nitric oxide pathway may not
be involved in the observed effect (Figure ). Moreover, TXA2 stimulator A23187 pretreatment
did not alter the analgesic effect of compound 6, which
indicated that compound 6 does not affect COX-1.
Figure 5
Graph showing
change in formalin-induced hyperalgesia in the presence
of compounds 2, 3, 4, 6, 10, 12, 17, 18, and 22. The values are given as mean ±
SD, * is p < 0.05 vs control.
Table 3
Analgesic Activity of Compounds 2, 3, 4, 6, 10, 12, 17, 18, and 22
compd
% inhibition of
algesia (analgesic effect)
2
57.41
3
55.56
4
61.34
6
68.24
10
54.85
12
74.25
17
69.02
18
58.95
22
61.94
Figure 6
Study of the mechanism of analgesic effect of compound 6 by using substance P, l-arginine, l-NAME, veratrine,
and calcium ionophore. All values are given as mean ± SD, # is p < 0.05 vs compound 6.
Graph showing
change in formalin-induced n class="Disease">hyperalgesia in the presence
of compounds 2, 3, 4, 6, 10, 12, 17, 18, and 22. The values are given as mean ±
SD, * is p < 0.05 vs control.
Study of the mechanism of analgesic effect of compound 6 by using substance P, n class="Chemical">l-arginine, l-NAME, veratrine,
and calcium ionophore. All values are given as mean ± SD, # is p < 0.05 vs compound 6.
Anti-Inflammatory
Studies
A marked
increase in the paw thickness of the control animals was observed
on carrageenan injection, and it was reached maxin class="Gene">mum by 30–45
min of injection. Treatment of the animals with indomethacin and compounds 6 and 12 was found to decrease the paw thickness
significantly as compared to the untreated control group (Figure ). In our experiments,
decrease of 18% (30 min) and 14% (60 min) in paw thickness at peak
response time (30–60 min) was noticed in the presence of compound 6, whereas the presence of indomethacindecreased inflammation
by 30% (30 min) and 25% (60 min) (Figure ).
Figure 7
Effect of compounds 6 and 12 on carrageenan-induced
paw inflammation in mice. The values are taken as mean ± SEM.
*p < 0.05 vs control group.
Effect of compounds 6 and 12 on carrageenan-induced
n class="Disease">paw inflammation in mice. The values are taken as mean ± SEM.
*p < 0.05 vs control group.
Acute Toxicity Studies
OECD guidelines
for the acute oral toxicity of 14 days dun class="Species">ration were followed for
checking the toxicity of compound 6.[21] Briefly, four groups comprising three animals in each group
were studied. Vehicle was given to the first group, whereas second,
third, and fourth groups were treated with compound 6 at 50, 300, and 2000 mg kg–1 dose. As per the
protocol of acute toxicity studies, there was no mortality or any
gross behavioral impairment after the 14th day. The tissue histology
revealed no significant lesions except for some degree of congestion
especially in the renal photomicrograph, and also the mesangium appeared
shrunken and increased capsular space was seen in comparison to the
control group (Figure ). Hence, desirably, compound 6 did not exhibit toxicity
even at a dose of 2000 mg kg–1.
Figure 8
Histology of liver: (A)
control and (B) compound 6 treated. Histology of kidney:
(C) control and (D) compound 6 treated. Histology of
myocardium: (E) control and (F) compound 6 treated. Each
image was 20× magnified.
Histology of liver: (A)
control and (B) compound 6 treated. Histology of kidney:
(C) control and (D) compound 6 treated. Histology of
myocardium: (E) control and (F) compound 6 treated. Each
image was 20× magnified.
In Vivo Pharmacokinetic
Studies
The
in vivo pharmacokinetic studies of compound 6 were carried
out by using a male wistar rat (250–300 g). The compound was
suspended in 0.1% critical n class="Species">micelle concentration (CMC) and administered
i.p. to the rats at a dose of 10 mg kg–1. The animals
were anesthetized with ketamine (50 mg kg–1 i.p.).
The blood samples were withdrawn from the jugular vein at an interval
of 30 and 60 min and 2, 3, 4, 6, 8, 11, and 24 h of compound administration.
The concentration of compound in the serum was determined using liquid
chromatography–mass spectrometry (LC–MS). The different
pharmacokinetic parameters were determined (Table , Figure ) following non compartmental analysis in PK solver
software. Compound 6 exhibited half-life 5.5 h and Cmax 58.5 μg mL–1.
Table 4
Pharmacokinetic Studies of Compound 6
parameter
unit
value
lambda_z
1/min
0.002110056
t1/2
Min
328.497015
Tmax
Min
180
Cmax
μg/mL
58.5
AUC 0–t
μg/mL·min
11 421.9
MRT 0–inf_obs
Min
489.609039
Vz/F_obs
(mg)/(μg/mL)
0.379638448
Cl/F_obs
(mg)/(μg/mL)/min
0.000801058
Figure 9
Pharmacokinetic
study of compound 6.
Pharmacokinetic
study of compound 6.
Isothermal
Titration Calorimetric (ITC) and
UV–Vis Experiments
Because the biological results
were in favor of COX-2 as the cellular target of compound 6, the binding affinity of the compound with the enzyme was checked
with ITC experiments. Isothermal titration calorimetry measures the
magnitude of the two thermodynamic terms: the enthalpy (ΔH) and entropy (ΔS) change in a single
experiment and the combination of these two parameters defines the
binding affinity (binding constant, K) between the
two chemical entities. The solution of enzyme in phosphate buffer
at pH 7.4 was put in the sample cell, and the solution of compound 6 (in the syringe) was injected stepwise (2 μL of 50
μM) after an interval of 120 s. For one experiment, 19 consecutive
additions of the compound were made, and Ka, ΔH, and ΔS were measured. The heat
change Q involved in the active cell during the interaction
of the compound and the enzyme is given by eq Here, Mt denotes
the total concentration of the enzyme, n represents
the binding sites of the enzyme, Vo is
the cell volume, and ΔH is the molar heat of
ligand binding.For the ith injection of the
compound with volume dV to the cell-containing enzyme, the enthalpy change ΔH(i) is given by eq and the values of different parameters are
given in Table .
Table 5
Isothermal
Calorimetric Data of Compound 6 during Its Interaction
with COX-1 and COX-2
The negative values of free energy (ΔG =
−33.41 kJ/mol) and enthalpy (ΔH = −40.52
kJ/mol) indicated that the binding of compound 6 with
COX-2 is spontaneous and exothermic.UV–vis spectral
studies with the solution of compound 6 and COX-2 were
also performed, and the binding constant
of the compound with n class="Gene">COX-2 was calculated using the Benesi–Hildebrand
equation.where Af is the
absorbance of the free host, Aobs is the
absorbance observed, Afc is the absorbance
at saturation, K is the binding constant, and L is the ligand concentration. Compound 6 showed
significant interaction with COX-2 with a binding constant, Ka 1.92 × 105 M–1.
Docking Studies and Molecular Dynamics (MD)
Simulation
Molecular docking of the compounds in the enzyme
active site was performed so that the nature of interactions between
the compounds and the active site amino acid residues of the enzyme
is explored. A 2.4 Å resolution structure of ovine COX-2 in complex
with AA (n class="Disease">PDB ID 1CVU)[22] and a 3.0 Å resolution structure
of COX-1 in complex with AA (PDB ID1DIY)[23] were taken
from the protein data bank (www.rcsb.org). After refinement, these proteins were used for performing molecular
docking and MD experiments. Flexibly docking mode of Schrodinger software
package[24] was used for molecular docking,
and the nature of interactions of compounds with the surrounding amino
acids was parameterized by using the binding modes and binding affinities
in the active site of COX-2. Docking procedure was validated by docking
AA in the active site of COX-2, and its root-mean-square deviation
(rmsd) from the native AA was 1.18 Å (Figure S152). Compound 6 exhibited two H-bond interactions
through its carbonyl oxygen and one of the three nitrogens of triazine
with S530 (2.25 Å) and R120 (2.79 Å)—the amino acids
which play major role during the catalytic phase of COX-2 (Figure a). It also displayed
π–π interactions with Y355 and cation−π
interaction with R513 (Figure a,b). Compounds 3, 4, 8, 10, 11, 12, 13, 17, 18, 19, and 22 also showed well-docked poses in the COX-2 active site
(Figures S153–S198, Supporting Information) exhibiting H-bond and hydrophobic interactions. However, compounds 9 and 14 did not dock in the active site pocket
of COX-2.
Figure 10
(a) Crystal coordinates of compound 6 in association
with the COX-2 active site (PDB ID 1CVU): (A) 3D view and (B) 2D view. (b) Hydrophobic
interactions of compound 6 with the active site residues
of COX-2.
(a) Crystal coordinates of compound 6 in association
with the COX-2 active site (n class="Disease">PDB ID 1CVU): (A) 3D view and (B) 2D view. (b) Hydrophobic
interactions of compound 6 with the active site residues
of COX-2.
To elucidate the importance of
protein flexibility in the ligand
binding site and to observe the dynamics of protein–ligand
interactions, the energy-minimized docked complex of compound 6 with COX-2 was subjected to MD simulations in an aqueous
solution environment for 50 ns. The overall stability of the system
under simulation was evaluated using the rmsd of the backbone atoms.
It was found that the rmsd of the protein backbone is significantly
stable over the course of the MD simulation (Figure ). The n class="Chemical">oxygen atom of morpholine and the
nitrogen atom of the triazine ring of compound 6 effectively
interact with the surrounding water molecules, which in turn are involved
in the formation of hydrogen bonds with E524 and R513 (Figure ). The stacked bar charts
were normalized over the course of the trajectory. V523 formed strong
hydrophobic interactions with compound 6, which were
conserved along the simulations. Compound 6 formed hydrogen
bonds and hydrophobic interactions with Y355 and L352 for more than
50% of the simulation time (Figure ). N87 also contributed in H-bonding and water bridges
with the ligand. Moreover, H90, P86, Y385, F518, and A527 exhibited
polar and nonpolar interactions for <50% of the MD time.
Figure 11
Rmsd of backbone
atoms during evolution of trajectory of Cα
(blue), side chain of protein (brown), and atoms of ligand 6 (pink) are shown.
Figure 12
Interactions of compound 6 with the protein residues
during evolution of trajectory (0.00–50 ns) that occur more
than 10%.
Figure 13
Interaction analysis between protein
and compound 6 throughout the simulation over the period
0.00–50 ns. The
stacked bar charts were normalized throughout the trajectory.
Rmsd of backbone
atoms during evolution of trajectory of Cα
(blue), side chain of protein (brown), and atoms of ligand 6 (pink) are shown.Interactions of compound 6 with the protein residues
during evolution of trajectory (0.00–50 ns) that occur more
than 10%.Interaction analysis between protein
and compound 6 throughout the simulation over the period
0.00–50 ns. The
stacked bar charts were normalized throughout the trajectory.Compound 10 also
showed well-docked pose in the COX-2
active site. H-bonding interactions through one of the n class="Chemical">nitrogens of
triazine with phenolic −OH of Y355 side chain (2.23 Å)
and another through its carbonyl oxygen with S530 (2.35 Å) were
observed (Figure A). It also exhibited π–π interactions between
1,3,5-triazine ring and Y355 and between the aromatic ring of oxindole
part and W387 and Y385 of COX-2 (Figure A,B). The aromatic ring of indole was also
involved in π–π interactions with Y355 and F518
of COX-2.
Figure 14
(A) Crystal coordinates of compound 10 and COX-2 active
site (PDB ID 1CVU). (B) Interactions of compound 10 with the amino acids.
(A) Crystal coordinates of compound 10 and COX-2 active
site (n class="Disease">PDB ID 1CVU). (B) Interactions of compound 10 with the amino acids.
NMR Experiments
for Enzyme–Ligand Interactions
To corroborate the
results of docking studies, solution-phase NMR
experiments of compound 10 in the absence and presence
of n class="Gene">COX-2 were performed. 1H chemical shift and spin–lattice
relaxation time (T1) of various protons
were measured by 1D1H NMR and inversion recovery NMR experiments.
The 1H NMR spectrum of 10 (12 mM) was recorded
in 0.5 mL of DMSO-d6 at 25 °C (blue
trace, Figure ).
Addition of 5 μL of COX-2 to the solution of compound 10 resulted in significant upfield shift of aromatic protons
H-5, H-9, H-4, and H-16 (Figure ), whereas no visible change in the CH2 protons
of morpholine and piperidine ring was observed. These observations
indicated that compound 10 interacts with COX-2. The
protons H-5, H-9, H-4, and H-16 are probably under the shielding effect
of hydrophobic residues of COX-2. The binding of compound with the
enzyme was also confirmed by T1 experiments
where spin–lattice relaxation time (T1) of various protons of compound 10 was measured
(Figure ). Characteristically,
supporting the chemical shifts data, the relaxation time of H-5, H-9,
H-1, H-4, and H-16 protons was considerably decreased in the presence
of COX-2. Hence, the results of T1 measurement experiments
indicated that an aromatic ring of indole and oxindole part of molecule 10 interacts with COX-2. All these protons were found to interact
with the amino acid residues during the docking of compound 10 in the active site of COX-2 (Figure ). Probably, these protons come under the
shielding effect of the hydrophobic residues of the enzyme, and this
was evident from the upfield shift of their resonance frequencies
in the presence of the enzyme. Because of the less solubility of compound 6 in comparison to the solubility of 10 for NMR
experiments, we did not perform 1H NMR and T1 experiments with 6.
Figure 15
Part of 1H NMR spectrum of compound 10 (blue
trace) and compound 10 in the presence of COX-2 (red
trace).
Figure 16
1H T1 relaxation times of
compound 10 (12 mM) in the absence (red dots) and presence
(blue dots) of COX-2.
Figure 17
Compound 10 was positioned in the interacting pocket
of COX-2 (PDB ID 1CVU). Arrows indicate the protons interacting with the amino acid residues.
The same protons undergo change in the chemical shift and T1 in the presence of COX-2.
Part of 1H NMR spectrum of compound 10 (blue
trace) and compound 10 in the presence of n class="Gene">COX-2 (red
trace).
1H T1 relaxation times of
compound 10 (12 mM) in the absence (red dots) and presence
(blue dots) of n class="Gene">COX-2.
Compound 10 was positioned in the interacting pocket
of COX-2 (n class="Disease">PDB ID 1CVU). Arrows indicate the protons interacting with the amino acid residues.
The same protons undergo change in the chemical shift and T1 in the presence of COX-2.
3D-Quantitative Structure–Activity
Relationship Study Model
Using partial least square (PLS)
factor of two, the field-based 3D-quantitative structure–activity
relationship (QSAR) model was genen class="Species">rated by correlating the activity
with steric, electrostatic, hydrophobic, hydrogen bond donor (HBD),
and hydrogen bond acceptor (HBA) factors (Figures and S205). Leave-one-out
cross-validation and non-cross-validation analysis gave R2CV and R2 0.66 and 0.96, respectively. The standard error of the estimate
was 0.18, and the F ratio was 102.0 (Tables and S3). It was observed that the steric, electrostatic, hydrophobic, HBA,
and HBD fields contribute 0.46, 0.093, 0.209, 0.227, and 0.00, respectively,
for the activity of the compound. The green region in Figure a represents that bulky substituents
are favorable for the activity, whereas the negative effect of the
steric substituents is represented by the yellow region. The diagrammatic
representation of the electrostatic factor is shown in Figure b where the blue contours
indicate electropositive groups that may increase the activity of
the compound and the red region displays electronegative groups. Desirably,
the electropositive region of compounds 6, 12, and 17 represented by Ph and CH3 groups
falls in the blue region, whereas the electronegative part of nitrogens
is placed in the red contours of the map. In compounds 16 and 19, the electronegative oxygen part of the molecules
lies in the unfavorable blue region, which might be leading to the
decrease in their activity. The placement of Cl-phenyl/2,6-dichlorophenyl
ring of compounds 3, 6, 9, 12, 18, and 22 in the yellow contours
might be responsible for the higher activity of these molecules (Figure c). The docked
complex of compound 6 with COX-2 (Figure b) also shows the hydrophobic interaction
through Cl-phenyl ring and hence complies to the comparative molecular
field analysis (CoMFA) model of the compound. The red color in Figure d favors the HBA
group/s, whereas the presence of HBA group/s in the magenta contours
is disfavored and may lead to reduced activity. In most of our compounds,
the nitrogen atom of triazine ring and carbonyl oxygen is involved
in hydrogen bonding with R120, S530, Y385, and Y355. Representing
HBD contribution, the blue-violet region in Figure e favors the activity of the compound, whereas
green color disfavors the activity. Because no HBD group is present
in the blue contours, the contribution of HBD in the present CoMFA
model is 0.00.
Figure 18
QSAR contour maps showing contribution of various descriptors:
(a) steric factor—yellow represents negative saturation and
green represents positive saturation; (b) electrostatic factor—blue
indicates positive effect and red represents negative saturation;
(c) hydrophobic contours—white region represents negative and
yellow represents positive saturation; (d) HBA—magenta region
shows negative effect and maroon indicates positive saturation; (e)
HBD—green represents negative saturation and blue-violet represents
positive saturation. The molecules with better activity are shown
in the form of a tube and inactive molecules in the form of thin wires.
Table 6
PLS-Based Statistical
Parameters for
the Selected 3D-QSAR Modela
PLS factors
SD
R2
R2CV
R2 scramble
stability
F
P
rmse
Q2
Pearson-R
1
0.4160
0.7322
0.5459
0.3537
0.955
32.8
9.49 × 10–5
0.38
0.4999
0.7701
2
0.1899
0.9688
0.6599
0.6149
0.798
102
7.92 × 10–8
0.27
0.7452
0.9163
SD, standard deviation of regression; R, squared value of R2 for the
regression; F, variance ratio. Large value of F indicates more statistical regression; P, significance level of variance ratio. Smaller value indicates a
greater degree of confidence; rmse, root-mean-square error; Q, squared value of Q2 for the
predicted activities; Pearson-R, Pearson-R value for the correlation between the predicted and observed
activity for the test set.
QSAR contour maps showing contribution of various descriptors:
(a) steric factor—yellow represents negative saturation and
green represents positive saturation; (b) electrostatic factor—blue
indicates positive effect and red represents negative saturation;
(c) hydrophobic contours—white region represents negative and
yellow represents positive saturation; (d) HBA—magenta region
shows negative effect and maroon indicates positive saturation; (e)
HBD—green represents negative saturation and blue-violet represents
positive saturation. The molecules with better activity are shown
in the form of a tube and inactive molecules in the form of thin wires.SD, standard deviation of regression; R, squared value of R2 for the
regression; F, variance ratio. Large value of F indicates more statistical regression; P, significance level of variance n class="Species">ratio. Smaller value indicates a
greater degree of confidence; rmse, root-mean-square error; Q, squared value of Q2 for the
predicted activities; Pearson-R, Pearson-R value for the correlation between the predicted and observed
activity for the test set.
Barring the HBD descriptor that was contributing equally in all
the compounds, the collective role of steric and hydrophobic parameters
is shown in Figure . The large green contour present at the center of Cl-phenyl ring
and 2,6-dichlorophenyl ring shows steric contribution to the activity
of compounds 3, 6, 9, 10, 11, 17, and 22,
whereas in compound 14, 2,6-dichlorophenyl ring did not
fall in the green region. For the contribution of hydrophobicity,
the favorable yellow region was found around the Cl-phenyl ring and
2,6-dichlorophenyl ring of compounds 3, 4, 6, 9, 10, 12, 17, and 22. Moreover, the hydrophobic
contours partially overlap with sterically favorable region. Therefore,
the COX-2 inhibitory activity of compounds 3, 4, 6, 9, 10, 12, 17, and 22 seems to be influenced by
the synergistic effect of steric and hydrophobic parameters.
Figure 19
Steric (A)
and hydrophobic (B) contour fields generated around
all compounds.
Steric (A)
and hydrophobic (B) contour fields generated around
all compounds.Significant fitness
between the experimental activity and the predicted
activity of the training and test set was observed in the QSAR model
(Table S3) (Figures and 21). A common
pharmacophore model was generated by aligning all the active and inactive
ligands to the pharmacophore (Figure ).
Figure 20
Scatter plot of observed versus predicted activity for
training
and test set compounds.
Figure 21
Plot between observed and predicted COX-2 inhibitory activity of
training set (a) and test set molecules (b) using a field-based 3D-QSAR
model.
Figure 22
Alignment of all ligands (active and
inactive) to the pharmacophore.
Scatter plot of observed versus predicted activity for
training
and test set compounds.Plot between observed and predicted COX-2 inhibitory activity of
training set (a) and test set molecules (b) using a field-based 3D-QSAR
model.Alignment of all ligands (active and
inactive) to the pharmacophore.
Conclusions
With the help of in vitro
and in vivo experiments on a series of
compounds, we were able to identify compound 6 as a new
lead for anti-inflammatory drugs. Compound 6 exhibited
IC50 20 nM and Ka 4.83 ×
105 M–1 for COX-2. The hydrophobic interactions
of 6 with the enzyme were apparent from the change in
chemical shifts and T1 of aromatic protons.
Significant analgesic and anti-inflammatory activity of compound 6 with minin class="Gene">mum toxicity risk was observed over Swiss albino
mice. We observed a close correlation between the results of solution-phase
NMR experiments and the molecular modeling studies, which could be
helpful for the further refinement of the structure of the molecule.
Experimental Section
General
Melting
points of the solid
compounds were determined in capillaries. Using CDCl3 and/or
n class="Chemical">dimethyl sulfoxide (DMSO)-d6 as the solvent
and tetramethylsilane as the internal reference, 1H and 13C NMR spectra were recorded on Bruker 500 and 125 MHz NMR
spectrometers. The chemical shifts are in parts per million, and coupling
constants are given in hertz. For representing the multiplicity of
signals in 1H NMR, s was used for singlet, d for doublet,
dd for double-doublet, t for triplet, and m for multiplet. A Bruker
micrOTOF Q II mass spectrometer was used for recording mass spectra.
All the compounds were procured as single geometrical isomer (Z-isomer)
except in the case of compounds 9 and 14 where inseparable E- and Z-isomers were obtained in a 2:1 ratio.
The purity of the compounds (P %) was assessed by
the q1H NMR method (absolute q1H NMR with internal
calibration),[25] and it was >98%.where MW is the molecular weight; P is the purity of the internal calibrant; mIC represents the amount of internal calibrant; ms is the amount of sample (compound); Int is
the integral; and n is the number of protons for
an NMR signal. IC is the internal calibrant, and t is the target analyte
or compound.
Synthesis of Compounds 1a and 1b
Cyanuric chloride (1 g, 5.43 mmol) was taken inn class="Chemical">acetone (40 mL) at 0 °C, and morpholine (0.94 g, 10.86 mmol)
was added dropwise followed by the addition of triethylamine (1.09
g, 10.86 mmol). Then, the reaction mixture was stirred at 25–28
°C for 1 h. The reaction was quenched with water and extracted
with ethyl acetate (4 × 25 mL). The organic layer was separated,
dried over Na2SO4, and concentrated in vacuum
to procure crude product that was column-chromatographed using ethyl
acetate and hexane as eluents to procure products 1a and 1b.
Synthesis of Compounds 3–6: General Procedure
Compounds 3–6 were
prepared through Knoevenagel condensation of compound 2 (1 mmol) with active methylene compounds including n class="Chemical">1-(3-chlorophenyl)-3-methyl-2-pyrazolin-5-one; N,N-dimethylbarbituric acid; and oxindole
and indolinone (1 mmol) in the presence of piperidine in CHCl3 at 100 °C for 2 h under microwave irradiation. The completion
of the reaction was monitored with thin-layer chromatography (TLC).
After completion of the reaction, the reaction mixture was quenched
with water and extracted with CHCl3 (4 × 25 mL). The
organic layer was separated, dried over Na2SO4, and concentrated in vacuum to procure crude product. The crude
product was further purified by diethyl ether or by recrystallization
in chloroform: methanol (2:8) to obtain compounds 3–6 with yield 80–85%.
Synthesis
of Compounds 9–12: General Procedure
Compounds 9–12 were
prepared through Knoevenagel condensation of compound 7 (1 mmol) with active methylene compounds including n class="Chemical">1-(3-chlorophenyl)-3-methyl-2-pyrazolin-5-one; N,N-dimethylbarbituric acid; and oxindole
and indolinone (1 mmol) in the presence of piperidine in CHCl3 at 100 °C for 2 h under microwave irradiation. After
completion of the reaction (TLC), the reaction mixture was quenched
with water and extracted with CHCl3 (4 × 25 mL). The
chloroform layers were collected, dried over Na2SO4, and concentrated in vacuum to procure crude product that
was purified by washing with ether or by recrystallization in chloroform:
methanol (2:8) to obtain compounds 9–12 with yield
70–85%.
Synthesis of Compounds 13–17: General Procedure
Compounds 13–17 were
prepared by the reaction of compound 7 (1 mmol) with
active methylene compounds including 1-(3-chlorophenyl)-3-methyl-2-pyrazolin-5-one; n class="Chemical">N,N-dimethylbarbituric acid; and oxindole
and indolinone (1 mmol) in the presence of pyrrolidine in CHCl3 at 100 °C for 2 h under microwave irradiation. After
the reaction is completed (TLC), the reaction mixture was quenched
with water and extracted with CHCl3 (4 × 25 mL). The
chloroform part was dried over Na2SO4 and concentrated
in vacuum to procure crude product. The crude product was further
purified by washing with ether or by recrystallization in chloroform:
methanol (2:8) to obtain compounds 13–17 with
yield 70–85%.
Synthesis of Compounds 18–20: General Procedure
A finely ground mixture
of 7 (1 mmol) and 1-(3-chlorophenyl)-3-methyl-2-pyrazolin-5-one/1,3-dimethyl
barbituric acid/n class="Chemical">oxindole (1 mmol) was heated at 155–160 °C
for 25–30 min. After completion of the reaction (TLC), the
crude reaction mass was washed with diethyl ether to procure pure
compounds 18–20.
2-Chloro-4,6-dimorpholino-1,3,5-triazine
(1a)
Compound 1a was procured as
per the procedure given above. Colorless solid, yield 65%, mp 175–176
°C. 1H NMR (500 MHz, n class="Chemical">CDCl3): δ 3.70–3.80
(m, 16H, 8 × CH2). 13C NMR (125 MHz, CDCl3): δ 43.86 (−ve, CH2), 66.53 (−ve,
CH2), 66.71 (−ve, CH2), 164.48 (ab, ArC),
169.69 (ab, ArC). HRMS (microTOF-QII, MS, ESI): calcd for C11H16N5O2Cl ([M + H]+),
286.1065; found, 286.1049.
2,4-Dichloro-6-morpholino-1,3,5-triazine
(1b)
Compound 1b was procured as
per the procedure given above. Colorless solid, yield 30%, mp 161–162
°C. 1H NMR (500 MHz, n class="Chemical">CDCl3): δ 3.76
(t, 4H, J = 9.82 Hz, 2 × CH2), 3.90
(t, 4H, J = 9.82 Hz, 2 × CH2). 13C NMR (125 MHz, CDCl3): δ 44.47 (−ve,
CH2), 66.38 (−ve, CH2), 164.10 (ab, ArC),
170.44 (ab, ArC). HRMS (ESI): calcd for C7H8N4OCl2 ([M + H]+), 235.0147; found,
235.0123.
NaH (1.2
mmol) was washed 3–4 times with dry n class="Chemical">hexane for the removal
of paraffin coating, and then it was suspended in dry ACN (20 mL).
Indole-3-carboxaldehyde (1 mmol) was added to NaH suspension in ACN,
and the reaction mixture was stirred at 0 °C for 5–10
min until the whole reactant gets dissolved. Then, compound 1a (1.2 mmol) was added with continuous stirring. On completion
of the reaction (TLC), the reaction was quenched by adding ice cold
water. The reaction mixture was extracted with ethyl acetate. After
drying over anhydrous Na2SO4, ethyl acetate
was removed under vacuum. The residue was column-chromatographed by
using ethyl acetate–hexane as eluents to procure pure product 2. Colorless solid, yield 76%, mp 244–245 °C.
IR (KBr): 3135, 2961, 2902, 2860, 1730, 1684, 1571 cm–1. 1H NMR (500 MHz, CDCl3): δ 3.82–3.95
(m, 16H, 8 × CH2), 7.41–7.44 (m, 2H, ArH),
8.35 (d, 1H, J = 7.18 Hz, ArH), 8.61 (d, 1H, J = 7.99 Hz, ArH), 8.84 (s, 1H, CH), 10.16 (s, 1H, CHO). 13C NMR (125 MHz, CDCl3): δ 43.93 (−ve,
CH2), 66.73 (−ve, CH2), 116.30 (+ve,
ArCH), 120.74 (ab, ArC), 122.02 (+ve, ArCH), 124.10 (+ve, ArCH), 125.29
(+ve, ArCH), 127.01 (ab, ArC), 136.33 (ab, ArC), 136.82 (+ve, ArCH),
162.74 (ab, ArC), 165.13 (ab, ArC), 185.82 (C=O). HRMS (ESI):
calcd for C20H22N6O3 ([M
+ H]+), 395.1826; found, 395.1819.
Piperidine (0.92 g, 10.86 mmol) was added
dropwise to the stirred solution of n class="Chemical">cyanuric chloride (1 g, 5.43 mmol)
in acetone (40 mL) at 0 °C, followed by the addition of triethylamine
(1.09 g, 10.86 mmol). Then, the reaction mass was stirred at rt for
1 h. The reaction was quenched with water and washed with ethyl acetate
(4 × 25 mL). The combined organic layers were dried over Na2SO4 and concentrated under vacuum to procure crude
product. The crude product was further purified by washing with diethyl
ether to obtain pure compound 23. White solid, yield
62%, mp 95–96 °C. 1H NMR (500 MHz, CDCl3): δ 1.58–1.59 (m, 8H, 4 × CH2), 1.65–1.66 (m, 8H, 4 × CH2), 3.73 (t, 8H, J = 5.28 Hz, 4 × CH2). 13C NMR
(125 MHz, CDCl3): δ 24.6 (−ve, CH2), 25.7 (−ve, CH2), 44.4 (−ve, CH2), 164.2 (ab, ArC), 169.5 (ab, ArC). HRMS (MS, ESI): calcd for C13H20N5Cl ([M + H]+), 282.1480;
found, 282.1508.
Synthesis of Compound 24
NaH (1.2 mmol) was washed with dry n class="Chemical">hexane and
suspended in dry
ACN (20 mL) to which indole-3-carboxaldehyde (1 mmol) was added at
0 °C. After stirring for 5–10 min, 1,3,5-triazine (1 mmol)
was added to the reaction mixture. On completion of the reaction (TLC),
ice cold water was added. The solid product was separated, which was
filtered and dried under vacuum to procure pure product 24. White solid, yield 45%, mp > 300 °C. 1H NMR
(500
MHz, CDCl3 + TFA): δ 7.58–7.61 (m, 3H, ArH),
7.65–7.68 (m, 3H, ArH), 8.39 (d, 3H, J = 7.66
Hz, ArH), 8.76 (d, 3H, J = 8.39 Hz, ArH), 9.11 (s,
3H, ArH), 10.16 (s, 3H, CHO). 13C NMR (125 MHz, CDCl3 + TFA): δ 110.8 (ab, C), 113.0 (ab, C), 115.3 (ab,
C), 116.2 (+ve, ArCH), 117.6 (ab, C), 122.3 (ab, C), 122.9 (+ve, ArCH),
126.5 (ab, C), 126.5 (+ve, ArCH), 127.3 (+ve, ArCH), 136.0 (ab, C),
138.4 (+ve, ArCH), 189.9 (C=O). HRMS (APCI): calcd for C30H18N6O3 ([M + H]+), 511.1513; found, 511.1521.
Docking
Procedure
Schrodinger software
package was used for performing the molecular docking of the compounds
in the COX-2 (n class="Disease">PDB ID 1CVU) and COX-1 (PDB ID1DIY) active site by following the procedure reported in the previous
paper.[20]
Field-Based
3D-QSAR Model
Field-based
QSAR tool of Schrodinger 2015-4 was used for the 3D-QSAR analysis
by following the reported procedure.[20]
Procedure for MD Simulation
The docked
complex of compound 6 with COX-2 was optimized using
MD simulation on DESMOND module in the Schrodinger Maestro 10.1 version
with On class="Disease">PLS-2005 force field in explicit solvent with the TIP3P water
model. The docked complex was placed in TIP3P water molecules adequately,
the dimensions of each orthorhombic water box were 10 Å ×
10 Å × 10 Å, and the systems were neutralized by adding
Na+ counterions to balance the net charges of the systems,
and then 0.15 M NaCl was added. The generated solvent model for the
docked complex contained about 59 176 atoms. Before MD simulations,
the systems were minimized and pre-equilibrated using the default
relaxation protocol executed in DESMOND. NVT MD simulations were performed
at 10 K for 100 ps with restraints on heavy atoms. Then, the system
was simulated for another 12 ps at 10 K with the same settings. This
was followed by NPT equilibration at 10 K for 12 ps. Then, the system
was simulated for 12 ps at 300 K with restraints on heavy atoms. Finally,
restraints on heavy atoms were removed, and the system was simulated
for 24 ps at 300 K with a thermostat relaxation time of 1 ps and a
barostat relaxation time of 2 ps. After minimization and equilibration,
MD simulations were performed at 300 K for 50 ns with the Martyna–Tobias–Klein
method. Data were collected every 1.5 ps during the MD runs.
Procedure for COX-1 and COX-2 Enzyme Immunoassay
COX-1/2
inhibitory immunoassays were performed by following the
already reported procedure.[20]
Human Whole Blood Assay
The human whole blood
assay was performed by using the
procedure given in the previous report.[20]
Analgesic and Anti-Inflammatory Activity
Prior permission was sought from the Institutional Animal Ethical
Committee of Guru Nanak Dev University, Amritsar, for using animals.
The analgesic and anti-inflammatory activity of the compounds was
determined by using male/female Swiss albino mice weighing 25–35
g. The animals were given free supply of food and n class="Chemical">water and kept at
22 ± 2 °C under 12 h light/dark cycle. A total of 12 groups
of animals with five animals in each group and a previously described[15b,19] procedure were used for screening the analgesic and anti-inflammatory
activity of the compounds.
Mechanistic Studies
Three groups
of mice, five in each group, were used to explore the mode of working
of the compound.[20] n class="Gene">Substance P, l-arginine, and NOS inhibitor, l-NAME pretreatment, were
given 30 min before administering compound 6.
Acute Toxicity Studies
Acute toxicity
studies were performed with four groups of animals, three animals
in each group. After 4 h of fasting, the first group of animals was
given vehicle and served as the control group; the second, third,
and the fourth groups were given compound 6 at doses
of 50, 300, and 2000 mg kg–1, respectively. During
the first 4 h, the animals were observed continuously and periodically
during the next 24 h. After 14 days, one animal each from the first
and third group was sacrificed, and the histological studies were
performed using H and E staining.
In Vivo
Pharmacokinetic Studies
The
in vivo pharmacokinetic properties were performed by using a male
Wistar rat (250–300 g). A dose of 10 mg kg–1 of 0.1% n class="Chemical">CMC suspension of the compound was administered i.p. to
the rats. The animals were anesthetized with ketamine (50 mg kg–1 i.p.). The blood samples from the jugular vein were
withdrawn in heparinized tubes at an interval of 30 and 60 min and
2, 3, 4, 6, 8, 11, and 24 h of compound administration. The samples
were centrifuged at 8000 rpm for 10 min at −4 °C, and
the clear serum was separated and stored at −20 °C until
analyzed. The concentration of compound in the serum was determined
using LC–MS after preparing the samples by the protein precipitation
method. Plasma sample (100 μL) was taken in 1.5 mL tube and
vortexed for 3 min. ACN (300 μL) with internal standard was
added to the above tube and vortexed for 5 min. The tube was centrifuged
at 4 °C, 16 000 rpm for 40 min. The compound with initial
concentration 3 mg mL–1 followed by serial dilution
was used for obtaining the standard curve. LC–MS was performed
with a Dionex ultimate 3000 HPLC system attached to a Bruker MicroTof
QII mass spectrometer. A 50 mm, 5 μm PRP C18 column was used
for high-performance liquid chromatography (HPLC), and the gradient
mobile phase consisted of water and acetonitrile (each containing
0.1% formic acid). The initial composition was 20% acetonitrile and
linearly increased to 100% in 30 min. The column eluent was introduced
to the ESI source of mass spectrometer operating in +ve mode. The
different pharmacokinetic parameters such as t1/2 (min), area under curve (AUC), Cmax (μg/mL), tmax(min), mean residence
time (min), and clearance (Cl) (mg)/(μg/mL)/min were determined
following noncompartmental analysis in PK solver software.[26]
NMR Studies for Protein–Ligand
Interactions
NMR experiments were performed on a Bruker AVANCE
500 NMR spectrometer
at 298.2 K. The diffusion measurements were carried out using “ledbpgppr2s”
pulse program. The gradient pulse length (P30) and
the diffusion time (D20) were kept fixed after optimizing
the parameters. The values for δ and Δ are 1.5 and 50
ms, respectively. A n class="Disease">longitudinal eddy current delay (D21) of 5 ms was used. The sample temperature was kept constant at
298 ± 0.1 K.
The pseudo-2D data were processed. The diffusion
coefficients were obtained with the help of the T1/T2 relaxation module of
Topspin as described in the diffusion manual of this software, whereas
the fitting-type “intensity” was used.
1H NMR T1 Relaxation
Time Measurements
1H NMR T1 relaxation time measurements were performed
on a Bruker AVANCE 500 NMR spectrometer at 298.2 K. The longitudinal
relaxation time (T1) was determined by
180°–90° inversion recovery pulse sequence. Sixteen
values of delay time (τ) were applied, and 16 scans for each
τ value were recorded. The preacquisition delay (D1) was set to 2 × T1 (5 s) of the
longest relaxation time. The value of the longitudinal relaxation
time was obtained with the help of T1/T2 relaxation module of Topspin as described
in the manual of this software, whereas the fitting function “invec”
and fitting type ‘‘area” were used.
UV–Visible Spectral Studies
Enzyme (3 μL)
was diluted to 100 μL in Tris-HCl buffer,
and 10 μM solution of the compound was prepared in HPLC-grade
n class="Chemical">DMSO and Tris-HCl buffer (1:9) (pH 7.4). Incremental addition of the
enzyme solution to the compound solution (10 μM) was made, and
spectra were recorded.
ITC Experiments
Solution (50 μM)
of compound 6 in HPLC-grade ethanol and n class="Chemical">phosphate buffer
(1:99) (pH 7.4) was prepared. Enzyme (10 μL) was diluted in
1 mL of phosphate buffer. The enzyme solution was taken in the sample
cell to which 19 additions of compound 6, each of 2 μL,
were made at equal intervals of 120 s. The experiments were performed
in triplicate, and the difference between the values was <1%. The
control experiments were performed with the same titrant (solution
of compound 6) with buffer in the cell. Also, the successive
buffer additions to the enzyme solution were carried out. The background
heat of the control experiments was subtracted from the main experiment.
To eliminate heat of mixing and heat of dilution, the total heat change Q involved in the cell volume Vo at fractional saturation Q after the ith injection was determined. The binding parameters were read with
origin 7.0 software of MicroCal, and a single binding site model was
used to fit the data.
Authors: Vladimir Garaj; Luca Puccetti; Giuseppe Fasolis; Jean-Yves Winum; Jean-Louis Montero; Andrea Scozzafava; Daniela Vullo; Alessio Innocenti; Claudiu T Supuran Journal: Bioorg Med Chem Lett Date: 2005-06-15 Impact factor: 2.823
Authors: Giovanna Li Petri; Maria Valeria Raimondi; Virginia Spanò; Ralph Holl; Paola Barraja; Alessandra Montalbano Journal: Top Curr Chem (Cham) Date: 2021-08-10
Figure 1
Scheme 1
Scheme 2
Scheme 3
Scheme 4
Scheme 5
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9
Figure 10
Figure 11
Figure 12
Figure 13
Figure 14
Figure 15
Figure 16
Figure 17
Figure 18
Figure 19
Figure 20
Figure 21
Figure 22