Discovery of N-pyridoyl-Δ2-pyrazolines as Hsp90 inhibitors

Article abstract of DOI:10.1002/ardp.201900192

Hsp90, as a key molecular chaperone, plays an important role in modulating the activity of many cell signaling proteins and is an attractive target for anticancer therapeutics. Herein, we report the discovery of N-pyridoyl-Δ²-pyrazoline analogs

Full text of DOI:10.1002/ardp.201900192

Received: 8 July 2019
Revised: 2 November 2019
Accepted: 17 November 2019
|
|
DOI: 10.1002/ardp.201900192
F U L L P A P E R
Discovery of N‐pyridoyl‐Δ2‐pyrazolines as Hsp90 inhibitors
Sundeep Kadasi1,2
Meghana Chary1
|
Ravali Yerroju1
|
Swetha Gaddam1
|
Nikhila Pullanagiri1
|
|
Divya Pingili3 Shiva Raj2
|
|
Nulgumnalli Manjunathaiah Raghavendra1,4
1Department of Pharmaceutical Chemistry,
Gokaraju Rangaraju College of Pharmacy,
Osmania University, Hyderabad, India
Abstract
Hsp90, as a key molecular chaperone, plays an important role in modulating the activity
of many cell signaling proteins and is an attractive target for anticancer therapeutics.
Herein, we report the discovery of N‐pyridoyl‐Δ2‐pyrazoline analogs as novel Hsp90
inhibitors by integrated approaches of drug design, organic synthesis, cell biology, and
qualitative proteomic analysis. Novel chemical compounds were designed and optimized
in the adenosine triphosphate‐binding site of Hsp90; lead optimized compounds were
found to have significant interactions with Asp93 and other amino acids crucial for
Hsp90 inhibition. The designed compounds were synthesized by a two‐step procedure;
different aromatic aldehydes were reacted with various acetophenones to form
substituted 1,3‐diphenyl‐prop‐2‐enones (Ic–Io), which upon reaction with isonicotinic
acid hydrazide in the presence of glacial acetic acid form N‐pyridoyl‐Δ2‐pyrazoline
compounds (PY1–PY13). Compounds PY3, PY2, and PY1 were identified as potential
leads amongst the series, with promising anticancer activity against human breast cancer
and melanoma cells, and the ability to inhibit Hsp90 similar to radicicol by drug‐affinity
responsive target stability proteomic analysis in a whole‐cell assay.
2Department of Chemistry, Osmania
University, Hyderabad, India
3Sri Venkateshwara College of Pharmacy,
Osmania University, Hyderabad, India
4Department of Pharmaceutical Chemistry,
Acharya & BM Reddy College of Pharmacy,
Bengaluru, India
Correspondence
Nulgumnalli M. Raghavendra, Department of
Pharmaceutical Chemistry, Acharya & BM
Reddy College of Pharmacy, Acharya Dr.
Sarvepalli Radhakrisnan Road,
Soladevanahalli, Chikkabanavara Post,
Bengaluru, Karnataka 560107, India.
Email: nmraghava@yahoo.in
Funding information
Department of Health Research,
Grant/Award Number: V.25011/283/2014‐
HRV.; Rajiv Gandhi University of Health
Sciences, Grant/Award Number: Advanced
Research Project 2019‐2020
K E Y W O R D S
anticancer activity, inhibitors, pyrazole, rational drug design
of malignancy.[6] In addition, Hsp90 has been identified as an important
extracellular mediator for tumor invasion,[7] and the expression of
|
1
INTRODUCTION
Hsp90 (heat shock protein 90) is a prevalent protein in mammalian cells.
Hsp90 functions as a molecular chaperone in the conformational
maturation, stability, and trafficking of several client proteins into their
biologically active forms.[1] Many Hsp90 client proteins are over-
expressed in cancer and are responsible for unrestricted cancer cell
proliferation and survival.[2] Prominent oncoproteins that are stabilized
and assisted by Hsp90 for oncogenesis are BRAF, Akt, Her2, cdk4, Src,
Flt‐3, hTert, c‐Met, Bcr‐Abl, and so forth.[3,4] Inhibition of Hsp90 results
in the simultaneous destabilization and degradation of multiple
oncogenic client proteins, leading to cancer cell growth inhibition and
apoptosis.[5] The pharmacologic blockade of the Hsp90 function is
claimed to have a combined inhibitory activity on all the hallmark traits
Hsp90 is also found to be amplified in cancer cells than normal cells.[8]
Thus, the discovery of Hsp90 inhibitors is considered an important
endeavor for anticancer drug development.
The complex natural product geldanamycin (GA) obtained from
Streptomyces hygroscopicus was the first natural Hsp90 inhibitor
reported.[9] Although too toxic to be developed as an anticancer
drug,[10] its optimization by semisynthesis resulted in two pro-
mising derivatives: 17‐allylamino‐17‐demethoxygeldanamycin (17‐AAG)
and 17‐(2‐dimethylamino)ethylamino‐17‐demethoxygeldanamycin (17‐
DMAG), which are more bioavailable than geldanamycin.[11,12] The
concern regarding the toxicity of geldanamycin analogs due to their
redox‐active quinone moiety[12,13] led to the discovery of similar
pharmacophoric compounds, such as macbecins[14] and herbimycins[15]
with potent anticancer activity, which are presently under clinical trial.
Radicicol (RAD), an antibiotic obtained from Chaetomium chiversii, is the
Abbreviations: DARTS, drug‐affinity responsive target selectivity; Hsp90, heat shock
protein 90; GA, geldanamycin; RAD, radicicol.
|
Arch Pharm Chem Life Sci. 2019;e1900192.
wileyonlinelibrary.com/journal/ardp
© 2019 Deutsche Pharmazeutische Gesellschaft
1 of 9
https://doi.org/10.1002/ardp.201900192

2 of 9
KADASI ET AL.
|
most potent inhibitor of Hsp90 in vitro, but the reactivity of its epoxide
group and the sensitivity of its conjugated double bonds to Michael
additions render it inactive in vivo.[16] Gedunin, a tetranortriterpenoid
isolated from the Indian neem tree (Azadirachta indica), which was
reported to have anticancer, antimalarial, and anti‐inflammatory
properties, is under clinical investigation.[17,18]
diverse in their chemical structure, they often bind to multiple proteins
due to their nonspecificity. Nonselective ATP‐binding ligands interact
with many proteins, exhibiting toxicity and leading to the failure of a
drug.[32,33] The N‐terminal domain of Hsp90 is homologous to the
members of the Hsp90 family, as well as to the members of the ATPase/
kinase GHKL superfamily.[34] Molecular design and optimization of a
selective N‐terminal domain Hsp90 inhibitor is a challenging and
complex process. Integrated approaches of bioinformatics, medicinal
chemistry, and polypharmacology are to be given importance in
developing Hsp90 inhibitors.
Experience with natural products generated interest in alternative
chemotypes. The purine class of compounds were the first synthetic
compounds discovered to have potent Hsp90 inhibition.[19] Using high‐
throughput screening,
a
novel 3,4‐diaryl pyrazole resorcinol
(CCT018159) was identified to have potent Hsp90 inhibition.[20] NVP‐
AUY922 (VER52296) is an isoxazole derivative initially developed by
optimization of a lead compound (CCT018159) and exhibits strong
anticancer activity against many mammalian cancer cells.[21] Computa-
tional approaches have produced many types of small‐molecule Hsp90
inhibitors, including pyrazoles,[22] resorcinol‐containing triazoles,[23]
isoindoles,[24] imidazoles,[25] indazol‐4‐ones,[26] and so forth. The
focused approach on the discovery of Hsp90 inhibitors credited
with many compounds in clinical trials such as NVP‐AUY922 (phase II;
Novartis),[27] ganetispib (STA‐9090, phase II; Synta),[28] XL‐888 (phase I;
Exelixis),[29] PU‐H71 (phase I; Memorial Sloan–Kettering Cancer
Center),[30] PF‐4929113/SNX‐5422 (phase I; Pfizer),[31] and so forth
(Figure 1). Though adenosine triphosphate (ATP)‐binding site ligands are
In continuation of our quest toward the discovery of small‐
molecule inhibitors of Hsp90,[35–38] in this article, we report the
molecular modeling, chemical synthesis, and biological evaluation of
N‐pyridoyl‐Δ2‐pyrazolines as Hsp90 inhibitors.
|
2
RESULTS AND DISCUSSION
|
2.1
Molecular modeling and docking simulations
We began our study by formulating a hypothesis to simulate
constraints that approximate better interactions in the ATP‐binding
site of Hsp90. The initial approach treats this binding site as rigid
FIGURE 1 Representative example of potential Hsp90 inhibitors of both natural and synthetic origin

KADASI ET AL.
3 of 9
|
while probing the conformational freedom of test ligands to establish
a complementary shape in the active site. Pyrazole scaffold inhibitors
of Hsp90 and insights gathered from the study of the ATP‐binding
site of Hsp90 motivated the selection of a trisubstituted pyrazoline
scaffold for targeting Hsp90 protein.[22,25,39,40] Active site prediction
by DoGSiteScorer reveals the ATP‐binding site as the best binding
pocket for Hsp90 protein (Table S1).[42] The validation of docking
methodology through the “pose selection” approach by superimpos-
ing the crystal ligand and the redocked ligand of Hsp90 protein
shows similar binding interactions, binding alignments, and satisfac-
tory RMSD between superimposed crystal ligand and redocked
ligand (Table S2).[42] Schrodinger's Glide XP docking[43] of pyrazoline
analogs in the rigid binding site of Hsp90 (PDB ID: 1YET)[44] endorse
PY series having significant binding interactions with Hsp90 protein
(Figure 2a–d). The detailed docking interactions are given in Table S3.
The structure–activity relationship of the lead molecules of PY series
was explored considering the binding interactions and their orienta-
tions in the ATP‐binding site of Hsp90, with natural Hsp90 inhibitor
GA and synthetic heterocyclic ligand “NVP‐AUY922” (Luminespib)
protein complexes. The N‐terminal ATP‐binding site of the
Hsp90–GA complex indicates that it is a 15 Å deep cone/pyramidal
shape pocket, 12 Å in diameter at the top surface and 8 Å wide
midway of the pocket.[44] The nature of the binding site is described
as a combination of polar and hydrophobic natures imparted by 17
amino acids lining the interior of the pocket, with increasing
hydrophobicity down to the bottom of the pocket, except induced
polarity (negative electrostatic potential) due to the amino acid
Asp93 and a polar residue Thr184, which are found to be crucial for
making ligand–protein polar‐bonding interactions. The surface of the
binding site is mostly polar (positive electrostatic potential) with
polar residues, Lys 58 and Lys 112, important for making polar
interactions with the ligand. The pyridyl moiety of PY3 is directed
toward the adenine‐binding pocket of the natural substrate ADP at
the bottom of the site, making crucial H‐bonding interactions with
amino acids Asp93 and Ser52 (Figure 2).[45] A similar interaction was
found with the carbamate residue of GA with Asp93, which is likely
to be a crucial interaction amongst the Hsp90 inhibitors. In addition,
the pyridyl “NH” of PY3 makes another important polar interaction
with Ser52, which was observed with the OH‐resorcinyl moiety of
NVP‐AUY922 and the carbamate of GA, suggesting that Asp93 and
Ser 52 are important residues for protein–ligand interactions.[21]
A
polar ‘NH’ of the heterocyclic ring is an essential pharmacophoric
feature for ligand binding in the design of Hsp90 inhibitors. The
central pyrazoline ring of PY3 overlaps the ribose sugar pocket of
bound ADP and the 5‐phenyl ring of PY3 is oriented toward Phe138,
creating hydrophobic stacking interactions. The 4″‐OH group of the
FIGURE 2 Binding interactions of PY1–PY3 in the adenosine triphosphate‐binding site (mesh or surface representation) of Hsp90 protein
(cartoon shape; PDB ID: 1YET). (a) H‐bonding interactions of green‐colored stick form PY3 with blue colored stick forms of Asp 93 (2.1 Å), Ser
52 (2.1 Å), Phe 138 (2.2 Å), and Gly 135 (2.0 Å). (b) H‐bonding interactions of cyano colored stick form PY1 with Asp 93 (1.8 Å) and Asn 51
(2.6 Å). (c) H‐bonding interactions of purple‐colored stick form PY2 with Thr 184 (2.0 Å), Phe 138 (2.1 Å) and Gly 137 (2.0 Å). (d) Overlap of stick
forms of PY3 (green color), PY1 (cyano color), and PY2 (purple color) in the active site of Hsp90 protein (cartoon shape; PDB code 1YET)

4 of 9
KADASI ET AL.
|
phenyl ring makes another H‐bonding interaction with Gly135,
stabilizing the complex, which is found in proximity to the planar
amide group of GA. Ligand PY1 showed hydrogen bonding interac-
tions with Asp93 and Asn51; the latter is in proximity to the Ser52,
making important H‐bonding interactions. Ligand PY3 shows a polar
H‐bonding interaction with Thr184, similar to carbamate of GA,
establishing that Asp93, Thr 184, Ser52 are very crucial for protein
1‐position of pyrazoline are crucial for nonpolar interactions with
hydrophobic residues of the protein. Heterocyclic pyrazoline is essential
for projecting the pharmacophoric features for essential interactions
with vital amino acids. To further evaluate the polar interactions in the
ATP‐binding site of Hsp90, docking simulations were performed with
different crystal complexes of Hsp90 protein available in the protein
data bank (PDB:ID 2BYH, 1OSF, and 4EGK).[39,46,47] Polar interactions of
the PY series are given in Table S4.
binding. Ligand PY2 interacts with Thr184 and
a stabilizing
interaction with amino acids Gly137 and Phe 138. Ligand PY4 shows
a slightly different orientation in the ADP‐binding site, without any
significant interaction with Asp93, Ser52, or Thr184, due to bulkiness
in its molecular structure and slightly greater rotational penalty (0.3),
yet making important hydrophobic interactions with amino acids
Val150, Leu107, Val136, Val186, Ala55, Met98, and ranks below the
lead molecules. PY5 interacts with Asp93 with an H‐bond parameter
of −0.9 and hydrophobic interactions with amino acids Ala55,
Phe138, Gly135, Val136, Leu107, suggesting it to be moderately
binding in the active site. PY6 interacts in a similar way with Asp93,
but the molecule is slightly low on the polar H‐bond parameter (−0.6),
suggesting it is a weakly binding ligand in docking analysis and
cytotoxicity study. Although ligand PY7 found to have a hydrogen
bond with Asp93, it has low electrostatic interactions and therefore
has low binding affinity than PY1, PY2, and PY3. PY8, PY9, PY10,
PY11, and PY13 do not show any significant interaction with a
crucial amino acid Asp93 and hence, are placed down the table in
terms of binding affinity, which corroborates with the cytotoxicity
assay. Although PY12 has significant interaction with Asp93, it is
ranked low in the cytotoxicity assay and docking analysis as the
|
2.2
CHEMISTRY
The synthetic protocol of N‐pyridoyl‐Δ2‐pyrazolines (depicted in
Scheme 1) began with a Claisen–Schmidt reaction between various
substituted benzaldehydes and substituted acetophenones to form 1,3‐
diphenyl‐prop‐2‐enone derivatives (Ic–Io). Purification of 1,3‐diphenyl‐
prop‐2‐enones was done by flash chromatography in the yields 72–85%.
Subsequently, the nucleophilic addition reaction of isonicotinic acid
hydrazide with the individual 1,3‐diphenyl‐prop‐2‐enones in glacial acetic
acid (as solvent), yielded cyclized pyrazolines (PY1–PY13) in 25–35%.
The structures of the final N‐pyridoyl‐Δ2‐pyrazoline compounds
(PY1–PY13) were confirmed by 1H‐NMR studies, and all the compounds
were found to have three characteristic peaks of a doublet of doublets
(dd) around 3.2–3.3, 3.7–3.8, and 5.6–5.8 ppm (parts per million), due to
Jabx coupling of protons on the pyrazoline ring. The infrared (IR) and mass
spectral data also confirmed the structural details of N‐pyridoyl‐Δ2‐
pyrazoline compounds (PY1–PY13). A single crystal of PY1 was obtained
by recrystallization from chloroform at room temperature (22 to 24°C).
X‐ray diffraction of PY1 was measured using radiation of wavelength
0.71073 Å at 296 K (Figure 3). The crystal system and space group were
found to be monoclinic and P2(1)/c (detailed crystallographic information
is given in Tables S5 and S6). The C5–N1 bond is found to be 1.288(3) Å,
representing unsaturation (C═N) as compared to C13‐N2 (saturated
bond, C–N), with a bond length of 1.387(3) Å, confirming it to be a
pyrazoline heterocycle. The C6–O1 bond length observed was 1.221 Å,
depicting a C═O system. The bond length of the C═N system (C9–N3
and C20–N3) for the pyridyl ring was found to be 1.316(4) and
1.323(4) Å. The bond length for N1 and N2 was observed to be
1.387(3) Å (data pertaining to the X‐ray diffraction studies of PY1 can be
molecule possesses greater rotational flexibility and
a greater
number of binding poses with a greater scope for nonselectivity.
The SAR analysis of pyrazolines alludes to the importance of a
trisubstituted pyrazoline scaffold with a heterocylic “NH” oriented
toward the bottom of the binding pocket (adenine‐binding site of ADP)
as an important pharmacophore. The presence of polar OH/OCH3 at the
3′,4′/3″,4″‐positions of the benzene ring contribute toward stabilizing
interactions of protein–ligand complex, increasing the polar and
electrostatic parameters for the drug‐like molecules, whereas electro-
negative atoms do not alter any binding affinity. Aromatic rings at the
3,5‐position of the pyrazoline and the heterocyclic aromatic ring at the
SCHEME 1 Synthesis of N‐pyridoyl‐Δ2‐pyrazoline analogs (PY1–PY13)

KADASI ET AL.
5 of 9
|
(MDA‐MB‐468).[49] The DARTS assay demonstrated that PY1–PY3
protect Hsp90β from protease digestion similarly to RAD (see the
1/1,000 pronase dilution in Figures 4 and 5). The protection of Hsp90
against proteolysis is possible only because of the stable Hsp90‐ligand
complex formed by PY compounds. RAD also protects the Hsp90 protein
against proteolysis. Untreated cell lysates failed to show the band of
Hsp90, as the Hsp90 was proteolyzed by pronase to fragments. The in
silico docking studies corroborate with the DARTS analysis, signifying
that PY1–PY3 interacts with Hsp90, and therefore shows promising
antiproliferative activity.
|
3
CONCLUSION
In conclusion, the drug design, chemical synthesis, and biological
evaluation of N‐pyridoyl‐Δ2‐pyrazolines are reported. The molecular
modeling and docking simulations predicted the binding potential of the
PY series in the N‐terminal ATP‐binding pocket of Hsp90. PY series
have crucial polar interactions with the important residues of Hsp90,
such as Asp93 and Thr184. In addition, nonpolar hydrophobic
interactions were also responsible for the efficient binding of
PY1–PY3 with Hsp90, which is evident by an admirable docking score
(−10.2 to −8.1 kcal/mol). Compounds PY1 and PY2 significantly reduced
the human breast cancer cell proliferation (IC50 1.60 and 2.8 μM),
whereas PY3 was moderate in action (IC50 12 μM). Compound PY1 was
also found to be a promising inhibitor of human melanoma cells (IC50
7.7 μM). Proteomic investigation in breast cancer cells implies the
Hsp90 binding property of PY1–PY3 molecules. All these investigations
propose N‐pyridoyl‐Δ2‐pyrazolines (PY1–PY3) as promising anticancer
compounds showing Hsp90 inhibition and show potential for further
development as effective anticancer agents.
FIGURE 3 Oak ridge thermal‐ellipsoid plot program (ORTEP)
diagram from X‐ray crystalographic study of PY1
accessed through the Cambridge Crystallographic Data Center [CCDC]
with the deposition number 1533986).
|
2.3
In vitro cell proliferation assay
To evaluate the biological activity of the PY series, anti-
proliferative activity screening by the 3‐(4,5‐dimethylthiazol‐2‐yl)‐2,5‐
diphenyltetrazolium bromide (MTT) assay method was performed against
human breast cancer cells MDA‐MB‐468 and human melanoma cells
A375.[48] The activities of the N‐pyridoyl‐Δ2‐pyrazoline compounds
(PY1–PY13) were compared with the reported Hsp90 inhibitor NVP‐
AUY922. PY1–PY3 turned out to be the promising compounds among
the series against breast and melanoma cancer cells (Table 1). The results
suggest that PY1–PY3 exhibited differential effects between cell lines of
human breast cancer (MDA‐MB‐468) and human melanoma (A375).
Compounds PY1–PY3 exhibited a robust cytotoxic effect against human
breast cancer with a range of IC50 values from 1.6 to 12 µM. Compounds
PY1–PY3 were also found to have significant cytotoxic activity against
human melanoma cells with a range of IC50 values from 7.7 to 22 µM.
|
4
EXPERIMENTAL
|
4.1
Molecular modeling and docking simulations
Molecular docking simulations were carried out on the Dell workstation
T1500 with the Windows 7 operating system using Schrodinger
Maestro 9.1 drug design software.[43] The Hsp90 protein bound with
GA (PDB ID: 1YET)[44] was downloaded from the RCSB Protein Data
Bank for drug design studies. The protein was prepared by filling
missing loops and missing side chains using a protein preparation wizard
application of the Schrodinger software. Chain A of Hsp90 protein was
further processed by removing nonreactive water molecules and crystal
ligand (GA). The ionized protein having the lowest penalty was energy‐
minimized using the optimized potential for liquid simulations 2005
force field incorporated in the Impref tool of Glide programme to finally
prepare processed 1YET protein. The grid was generated in the
processed protein by excluding the docked ligand in the active site using
a receptor grid generation tool of the Glide programme (the van der
Waals radius‐scaling factor was limited to 1.0 with a partial charge
cut‐off of 0.25). The active site was also predicted using DoGSiteScorer
|
2.4
Proteomic analysis
On the basis of the results of in vitro antiproliferative studies on human
cancer cells, we carried out the proteomic analysis of hit compounds
PY1–PY3 to explore the Hsp90 protein interaction. Drug‐affinity
responsive target stability (DARTS), a recent proteomics approach to
investigate small‐molecule binding to targets using protease‐based
digestion, was performed on the cell lysates of human breast cancer

6 of 9
KADASI ET AL.
|
TABLE 1 Physical properties and anticancer activity of N‐pyridoyl‐Δ2‐pyrazoline analogs (PY1–PY13)
IC50 (μM) SEMa
MDA‐MB‐468b
1.60 0.06
Compound
PY1
R1
R2
Yield (%)
55
m.p. (°C)
310
124
157
127
147
155
157
177
127
150
157
144
138
–
A375c
H
H
7.73 0.04
15.45 0.03
22.10 0.48
47.25 0.04
49.10 0.12
48.87 0.13
40.32 0.13
35.87 0.24
43.57 0.43
47.50 0.18
48.57 0.34
47.59 0.26
38.98 0.13
0.021 0.03
PY2
3′,4′‐OCH3
H
65
2.84 0.05
PY3
3′,4′‐OCH3
4″‐OH, 3″‐OCH3
4″‐OCH2C6H5, 3″‐OCH3
4″‐OCH2C6H5, 3″‐OCH3
3″,4″‐OCH3
3″,4″‐OCH3
4″‐OCH2C6H5
4″‐OCH2C6H5
4″‐Cl
73
12.05 0.03
45.80 0.06
45.10 0.03
35.72 0.03
34.75 0.15
38.58 0.16
46.87 0.23
48.68 0.14
44.35 0.34
37.59 0.43
36.68 0.37
0.012 0.01
PY4
2′,4′‐OCH3
60
PY5
H
65
PY6
3′,4′‐OCH3
75
PY7
4′‐Cl
74
PY8
2′,4′‐OCH3
66
PY9
H
70
PY10
PY11
PY12
PY13
NVP‐AUY922
H
75
2′,4′‐OCH3
4″‐Cl
68
2′,4′‐OCH3
4′‐OCH2C6H5(P)F
4″‐OCH2C6H5(P)F
–
73
H
75
–
–
Abbreviations: SEM, standard error of the mean; m.p., melting point.
aData represent the IC50 values for a 3‐day exposure to normalized to no drug controls and is the mean of triplicate experiments performed;
concentration ranges for PY1–PY3 compounds were 3.12 to 50 µM and for NVP‐AUY922 12.5 to 200 nM.
bHuman breast cancer.
cHuman melanoma.
software.[41] The ligands were subjected to Ligprep simulations to
generate energy‐minimized three‐dimensional structures (300 steps) by
investigating tautomeric, stereochemical, and ionization variations. The
ligprep out ligands were docked flexibly in the protein grid using Glide‐
extra precision (XP) simulations. Compounds having ≤300 atoms and
≤50 rotatable bonds were docked using five poses per ligand and
10,000 poses per docking run. Energies of residues within 12 Å of grid
were used for simulations. Poses having coulomb–vdW energy greater
than 0.0 kcal/mol and poses having an RMS deviation of 0.5 Å were
discarded.[42] Finally, the docked ligands were scored based on the
FIGURE 4 Proteomic analysis of RAD and PY3 mediated protection of Hsp90 in MDA‐MB‐468 lysate. One millimolar RAD and PY3 protect
Hsp90 from pronase degradation. Middle β‐actin blot demonstrates the compound effect is specific for Hsp90 because of the equivalent
proteolysis (+/−) compound. Bottom β‐actin is a loading control from nonproteolyzed samples run simultaneously on a separate gel.
Concentrations of pronase used are 1/10,000, 1/3,000, 1/1,000, and 1/300. DF, dilution factor; RAD, radicicol

KADASI ET AL.
7 of 9
|
FIGURE 5 Proteomic analysis of PY1 and PY2 analog mediated protection of Hsp90 in MDA‐MB‐468 cell lysate. One millimolar PY1 and
PY2 protect Hsp90 from degradation. The specificity of this effect is indicated by equivalent β‐actin degradation (+/−) compound (middle blot).
The bottom β‐actin blot acts as a loading control from the same sample but without proteolysis. Concentrations of pronase used are 1/10,000,
1/3,000, 1/1,000 and 1/300. DF, dilution factor
nonbonded interactions, such as lipophilic pair term, hydrogen bonding,
hydrophobic enclosure reward, and electrostatic rewards.
density of 1.0 × 104 cells/well in 80 μl and allowed to attach to the tissue
culture‐treated plastics after placing the 96‐well plate in a Nuncoverplate
for evaporation control for 4 hr. A five‐step, two‐fold drug dose dilution
series was prepared robotically in a Biomek 3000 (Beckman Coulter) in
the medium described previously. The test compounds were delivered as
20‐μl aliquots mixed into 80‐μl cell aliquot for a final exposure
concentration. A Day‐0 plate for each cell line was developed by the
robotic addition of 10 μl of 5 mg/ml MTT (in DMEM [Dulbecco’s modified
Eagle’s medium], low glucose, 0% fetal bovine serum, and without phenol
red) per well and allowed to develop for 4 hr after being placed in a
Nuncoverplate to control evaporation. Then, 100 μl of MTT solvent
(0.1 N HCl in anhydrous isopropanol with 10% Triton X‐100) was added,
the assay plates were tightly wrapped in foil and placed in sealed zip lock
bags to allow the solubilization of the MTT formazan product to occur
and progress to solubilization for reading in a spectrophotometer at
570 nm. The subtraction of background absorbance measured at 690 nm
was not performed. The Day‐3 test plates were developed as described
above. IC50 values were calculated using GraphPad Prism (Version 5.02,
GraphPad software).
Active site prediction for Hsp90, validation of docking methodol-
ogy, extra precision (XP) docking results of Maestro 9.1 Glide,
comparative docking interactions of the ligands in different crystal
structures of Hsp90 protein, and crystallographic information of PY1
are provided as Supporting Information.
|
4.2
Chemistry
1H‐NMR spectra were recorded in CDCl3 on a Bruker Avance 300 MHz
NMR spectrometer (Bruker BioSpin AG, Fallanden, Switzerland);
chemical shifts (δ) were reported in ppm with tetramethylsilane as an
internal standard. Mass fragmentation was recorded on an API2000 LC/
MS mass spectrometer (Bruker Daltonics Inc., Billerica, MA). X‐ray
diffraction studies were done using Bruker‐APEX III (X‐ray diffract-
ometer). Column chromatography was performed on Buchi flash
chromatography with C‐601 Pump Module, Pump Controller C‐610,
and Glass Column 26/230 cpl using silica gel (100–200 mesh). Infrared
spectra were obtained from FT‐IR‐Affinity‐1 spectrometer (Shimadzu,
Japan). Uncorrected melting points were determined on an electro-
thermal melting point apparatus. Unless otherwise noted, all solvents
and reagents were commercially available and used without further
purification. The spectroscopic characterizations and original spectra of
the synthesized compounds are provided in the Supporting Information.
The InChI data of the PY series with anticancer activity against MDA
MB468 and A375 cell lines are provided as Supporting Information.
|
4.3.2
Hsp90–small‐molecule inhibitor DARTs
assay using whole‐cell lysate[49]
MDA‐MB‐468 cells were lysed with lysis buffer (1 mM NaVO3, 50 mM
HEPES [4‐(2‐hydroxyethyl)‐1‐piperazineethanesulfonic acid], pH 7.4,
100 mM NaCl, 0.5% NP40, 1 mM EDTA [ethylenediaminetetraacetic
acid], 1 mM EGTA [ethylene glycol‐bis(β‐aminoethyl ether)‐N,N,N′,N′‐
tetraacetic acid], 50 μg/ml RNase, 1% Triton X‐100, 1% deoxycholic acid,
1 μg/1 μl leupeptin or Roche protease inhibitor mixture and 1× protease
mixture) for 15 min at 25°C. After centrifugation (13,200 rpm using
Eppendorf microcentrifuge 5415D; 15 min), the protein concentration of
the lysate was measured using MicroBCA™ protein assay kit. Twenty‐five
|
4.3
Biological activity
|
4.3.1
Cell antiproliferative activity assay
The growth inhibitory activity of the test compounds was determined by
MTT assay against human breast MDA‐MB‐468 and human melanoma
A375 cells.[48] Each cell line was seeded on 96‐well microplates at a

8 of 9
KADASI ET AL.
|
micrograms of protein cell lysate was incubated with 1 mM compound
and binding buffer (50 mM Tris HCl pH 8.0, 50 mM NaCl, 10 mM CaCl2)
to a 20‐μl final volume for 2 hr at room temperature. Samples were
digested with Pronase (Roche) at varying dilutions for 15 min at 25°C.
The reaction was stopped by the addition of 5 μl of 5× sodium dodecyl
sulfate (SDS) loading dye and this was immediately followed by boiling
samples at 95°C for 5 min. Samples were run in 4–15% gradient SDS‐
PAGE (SDS‐polyacrylamide gel electrophoresis) gels at 150 V for 60 min
followed by Western blot analysis. An equal aliquot of the nonproteo-
lyzed sample was run simultaneously on a separate gel to assess β‐acting
levels as a loading control. Blots were probed with anti‐Hsp90 antibody
(ADI‐SPA‐831; Enzo Life Sciences). The blots were stripped and reprobed
with anti‐β‐actin antibody (JLA20; DSHB University of Iowa) to
demonstrate that the β‐actin was proteolyzed equally in the presence
or absence of compound, and hence the compound‐mediated protection
of Hsp90 was specific. The same anti‐β‐actin antibody was used to probe
β‐actin levels in the loading control gel.
[9] L. Whitesell, E. G. Mimnaugh, B. de Costa, C. E. Myers, L. M. Neckers,
Proc. Natl. Acad. Sci. USA 1994, 91, 8324.
[10] J. G. Supko, R. L. Hickman, M. R. Grever, L. Malspeis, Cancer
Chemother. Pharmacol. 1995, 36, 305.
[11] R. C. Schnur, M. L. Corman, R. J. Gallaschun, B. A. Cooper, M. F. Dee,
J. L. Doty, M. L. Muzzi, J. D. Moyer, C. I. DiOrio, E. G. Barbacci, P. E.
Miller, A. T. O’Brien, M. J. Morin, B. A. Foster, V. A. Pollack,
D. Savage, D. E. Sloan, L. R. Pustilnik, M. P. Moyer, J. Med. Chem.
1995, 38, 3806.
[12] E. Sausville, J. Tomaszewski, P. Ivy, Curr. Cancer Drug Targets 2003, 3,
377.
[13] J. L. Eiseman, J. Lan, T. F. Lagattuta, D. R. Hamburger, E. Joseph, J. M.
Covey, M. J. Egorin, Cancer Chemother. Pharmacol. 2005, 55, 21.
[14] S. Tanida, T. Hasegawa, E. Higashide, J. Antibiot. 1980, 33, 199.
[15] K. Shibata, S. Satsumabayashi, A. Nakagawa, S. Omura, J. Antibiot.
1986, 39, 1630.
[16] N. Winssinger, S. Barluenga, Chem. Commun. (Camb) 2007, 1, 22.
[17] G. E. L. Brandt, M. D. Schmidt, T. E. Prisinzano, B. S. J. Blagg, J. Med.
Chem. 2008, 51, 6495.
[18] P. V. Borges, K. H. Moret, N. M. Raghavendra, T. E. Maramaldo
Costa, A. P. Monteiro, A. B. Carneiro, P. Pacheco, J. R. Temerozo,
D. C. Bou‐Habib, M. das Graças Henriques, C. Penido, Pharmacol. Res.
2017, 115, 65.
[19] G. Chiosis, M. N. Timaul, B. Lucas, P. N. Munster, F. F. Zheng, L. Sepp‐
Lorenzino, N. Rosen, Chem. Biol. 2001, 8, 289.
ACKNOWLEDGMENTS
[20] M. G. Rowlands, Y. M. Newbatt, C. Prodromou, L. H. Pearl,
P. Workman, W. Aherne, Anal. Biochem. 2004, 327, 176.
[21] P. A. Brough, W. Aherne, X. Barril, J. Borgognoni, K. Boxall, J. E.
Cansfield, K. M. J. Cheung, I. Collins, N. G. M. Davies, M. J. Drysdale,
B. Dymock, S. A. Eccles, H. Finch, A. Fink, A. Hayes, R. Howes, R. E.
Hubbard, K. James, A. M. Jordan, A. Lockie, V. Martins, A. Massey,
T. P. Matthews, E. McDonald, C. J. Northfield, L. H. Pearl,
C. Prodromou, S. Ray, F. I. Raynaud, S. D. Roughley, S. Y. Sharp,
A. Surgenor, D. L. Walmsley, P. Webb, M. Wood, P. Workman,
L. Wright, J. Med. Chem. 2008, 51, 196.
We are thankful to the Department of Health Research (V.25011/283/
2014‐HR), Ministry of Health & Family Welfare, Government of India,
and the Rajiv Gandhi University of Health Sciences, Karnataka
(Advanced Research Project 2019–2020), India, for providing financial
and moral assistance to this project. We thank Dr. Vincent Jo Davisson,
Dr. Tony Hazbun, and Mr. Raymond Fatig of the MCMP Department,
Purdue University, USA, for their constructive suggestions and support
throughout the project.
[22] E. McDonald, K. Jones, P. Brough, M. Drysdale, P. Workman, Curr.
Top. Med. Chem. 2006, 6, 1193.
[23] T. Y. Lin, M. Bear, Z. Du, K. P. Foley, W. Ying, J. Barsoum, C. London,
Exp. Hematol. 2008, 36, 1266.
CONFLICT OF INTERESTS
[24] A. J. Woodhead, H. Angove, M. G. Carr, G. Chessari, M. Congreve, J.
E. Coyle, J. Cosme, B. Graham, P. J. Day, R. Downham, L. Fazal,
R. Feltell, E. Figueroa, M. Frederickson, J. Lewis, R. McMenamin,
C. W. Murray, M. A. O’Brien, L. Parra, S. Patel, T. Phillips, D. C. Rees,
S. Rich, D. M. Smith, G. Trewartha, M. Vinkovic, B. Williams,
A. J. A. Woolford, J. Med. Chem. 2010, 53, 5956.
The authors declare that there are no conflict of interests.
ORCID
[25] R. Bao, C. J. Lai, H. Qu, D. Wang, L. Yin, B. Zifcak, R. Atoyan, J. Wang,
M. Samson, J. Forrester, S. Della Rocca, G. X. Xu, X. Tao, H. X. Zhai,
X. Cai, C. Qian, Cancer Res. 2009, 15, 4046.
Nulgumnalli Manjunathaiah Raghavendra
http://orcid.org/0000-
[26] K. H. Huang, J. M. Veal, R. P. Fadden, J. W. Rice, J. Eaves, J. P. Strachan,
A. F. Barabasz, B. E. Foley, T. E. Barta, W. Ma, M. A. Silinski, M. Hu, J. M.
Partridge, A. Scott, L. G. DuBois, T. Freed, P. M. Steed, A. J. Ommen,
E. D. Smith, P. F. Hughes, A. R. Woodward, G. J. Hanson, W. S. McCall,
C. J. Markworth, L. Hinkley, M. Jenks, L. Geng, M. Lewis, J. Otto,
B. Pronk, K. Verleysen, S. E. Hall, J. Med. Chem. 2009, 52, 4288.
[27] J. C. Bendell, T. M. Bauer, R. Lamar, M. Joseph, W. Penley,
D. S. Thompson, D. R. Spigel, R. Owera, C. M. Lane, C. Earwood,
H. A. Burris, Cancer Invest. 2016, 34, 265.
0002-0846-012X
REFERENCES
[1] L. H. Pearl, C. Prodromou, Annu. Rev. Biochem. 2006, 75, 271.
[2] D. Hanahan, R. A. Weinberg, Cell 2000, 100, 57.
[3] L. Neckers, P. Workman, Clin. Cancer Res. 2012, 18, 64.
[4] B. S. J. Blagg, T. D. Kerr, Med. Res. Rev. 2005, 26, 310.
[5] A. Maloney, P. Workman, Expert Opin. Biol. Ther. 2002, 2, 3.
[6] P. Workman, Curr. Cancer Drug Targets 2003, 3, 297.
[28] W. Ying, Z. Du, L. Sun, K. P. Foley, D. A. Proia, R. K. Blackman,
D. Zhou, T. Inoue, N. Tatsuta, J. Sang, S. Ye, J. Acquaviva, L. S. Ogawa,
Y. Wada, J. Barsoum, K. Koya, Mol. Cancer Ther. 2012, 11, 475.
[29] W. Xu, L. Neckers, Clin. Cancer Res. 2007, 13, 1625.
[30] R. M. Immormino, Y. Kang, G. Chiosis, D. T. Gewirth, J. Med. Chem.
2006, 49, 4953.
[7] B. K. Eustace, T. Sakurai, J. K. Stewart, D. Yimlamai, C. Unger, C.
Zehetmeier, B. Lain, C. Torella, S. W. Henning, G. Beste, B. T.
Scroggins, L. Neckers, L. L. Ilag, D. G. Jay, Nat. Cell Biol. 2004, 6, 507.
[8] Q. Cheng, J. T. Chang, J. Geradts, L. M. Neckers, T. Haystead, N. L.
Spector, H. K. Lyerly, Breast Cancer Res. 2012, 14, R62.
[31] P. Fadden, K. H. Huang, J. M. Veal, P. M. Steed, A. F. Barabasz,
B. Foley, M. Hu, J. M. Partridge, J. Rice, A. Scott, L. G. Dubois,

KADASI ET AL.
9 of 9
|
T. A. Freed, M. A. R. Silinski, T. E. Barta, P. F. Hughes, A. Ommen,
[44] C. E. Stebbins, A. A. Russo, C. Schneider, N. Rosen, F. U. Hartl,
N. P. Pavletich, Cell 1997, 89, 239.
W. Ma, E. D. Smith, A. W. Spangenberg, J. Eaves, G. J. Hanson,
L. Hinkley, M. Jenks, M. Lewis, J. Otto, G. J. Pronk, K. Verleysen,
T. A. Haystead, S. E. Hall, Chem. Biol. 2010, 17, 686.
[45] W. M. J. Obermann, H. Sondermann, A. A. Russo, N. P. Pavletich,
F. U. Hartl, J. Cell Biol. 1998, 143, 901.
[32] T. Force, D. S. Krause, R. A. van Etten, Nat. Rev. Cancer 2007, 7,
332.
[46] J. M. Jez, J. C. H. Chen, G. Rastelli, R. M. Stroud, D. V. Santi, Chem.
Biol. 2003, 10, 361.
[33] B. B. Hasinoff, D. Patel, K. A. O’Hara, Mol. Pharmacol. 2008, 74, 1722.
[34] C. Prodromou, L. Pearl, Curr. Cancer Drug Targets 2003, 3, 301.
[35] S. Dutta Gupta, M. K. Bommaka, G. I. Mazaira, M. D. Galigniana,
C. V. S. Subrahmanyam, N. L. Gowrishankar, N. M. Raghavendra, Int.
J. Biol. Macromol. 2015, 80, 253.
[47] C. Austin, S. N. Pettit, S. K. Magnolo, J. Sanvoisin, W. Chen,
S. P. Wood, L. D. Freeman, R. J. Pengelly, D. E. Hughes, J. Biomol.
Screen. 2012, 17, 868.
[48] M. V. Berridge, P. M. Herst, A. S. Tan, Biotechnol. Annu. Rev. 2005, 11,
127.
[36] S. Dutta Gupta, B. Revathi, G. I. Mazaira, M. D. Galigniana,
C. V. S. Subrahmanyam, N. L. Gowrishankar, N. M. Raghavendra,
Bioorg. Chem. 2015, 59, 97.
[49] B. Lomenick, R. Hao, N. Jonai, R. M. Chin, M. Aghajan, S. Warburton,
J. Wang, R. P. Wu, F. Gomez, J. A. Loo, J. A. Wohlschlegel,
T. M. Vondriska, J. Pelletier, H. R. Herschman, J. Clardy, C. F. Clarke,
J. Huang, Proc. Natl. Acad. Sci. USA 2009, 106, 21984.
[37] S. Dutta Gupta, D. Snigdha, G. I. Mazaira, M. D. Galigniana,
C. V. S. Subrahmanyam, N. L. Gowrishankar, N. M. Raghavendra,
Biomed. Pharmacother. 2014, 68, 369.
SUPPORTING INFORMATION
[38] S. Kadasi, T. E. M. M. Costa, N. Arukala, M. Toshakani, C. Duggineti,
S. Thota, S. D. Gupta, S. Raj, C. Penido, M. G. Henriques,
N. M. Raghavendra, Med. Chem. 2018, 14, 44.
Additional supporting information may be found online in the
Supporting Information section.
[39] P. A. Brough, X. Barril, M. Beswick, B. W. Dymock, M. J. Drysdale,
L. Wright, K. Grant, A. Massey, A. Surgenor, P. Workman, Bioorg.
Med. Chem. Lett. 2005, 15, 5197.
[40] Y. Yang, H. Liu, J. Du, J. Qin, X. Yao, J. Mol. Model. 2011, 17, 3241.
[41] A. Volkamer, D. Kuhn, F. Rippmann, M. Rarey, Bioinformatics 2012, 28,
2074.
How to cite this article: Kadasi S, Yerroju R, Gaddam S, et al.
Discovery of N‐pyridoyl‐Δ2‐pyrazolines as Hsp90 inhibitors.
Arch Pharm Chem Life Sci. 2019;e1900192.
[42] K. E. Hevener, W. Zhao, D. M. Ball, K. Babaoglu, J. Qi, S. W. White,
R. E. Lee, J. Chem. Inf. Model. 2009, 49, 444.
https://doi.org/10.1002/ardp.201900192
[43] Maestro 9.1, Schrödinger, LLC, New York, NY; 2011.