Tubulin-Binding 3,5-Bis(styryl)pyrazoles as Lead Compounds for the Treatment of Castration-Resistant Prostate Cancer

Article abstract of DOI:10.1124/MOL.119.118539

The microtubule-binding taxanes, docetaxel and cabazitaxel, are administered intravenously for the treatment of castration-resistant prostate cancer (CRPC) as the oral administration of these drugs is largely hampered by their low and highly variable bioavailabilities. Using a simple, rapid, and environmentally friendly microwave-assisted protocol, we have synthesized a number of 3,5-bis(styryl)pyrazoles 2a-l, thus allowing for their screening for antiproliferative activity in the androgen-independent PC3 prostate cancer cell line. Surprisingly, two of these structurally simple 3,5-bis(styryl)pyrazoles (2a and 2l) had concentrations which gave 50percent of the maximal inhibition of cell proliferation (GI₅₀) in the low micromolar range in the PC3 cell line and were thus selected for extensive further biologic evaluation (apoptosis and cell cycle analysis, and effects on tubulin and microtubules). Our findings from these studies show that 3,5-bis[(1E)-2(2,6-dichlorophenyl)ethenyl]-1H-pyrazole 2l 1) caused significant effects on the cell cycle in PC3 cells, with the vast majority of treated cells in the G₂/M phase (89percent); 2) induces cell death in PC3 cells even after the removal of the compound; 3) binds to tubulin [dissociation constant (K_d) 0.4 6 0.1 mM] and inhibits tubulin polymerization in vitro; 4) had no effect upon the polymerization of the bacterial cell division protein FtsZ (a homolog of tubulin); 5) is competitive with paclitaxel for binding to tubulin but not with vinblastine, crocin, or colchicine; and 6) leads to microtubule depolymerization in PC3 cells. Taken together, these results suggest that 3,5-bis(styryl)pyrazoles warrant further investigation as lead compounds for the treatment of CRPC.

Full text of DOI:10.1124/MOL.119.118539

Molecular Pharmacology Fast Forward. Published on April 2, 2020 as DOI: 10.1124/mol.119.118539
This article has not been copyedited and formatted. The final version may differ from this version.
MOL # 118539
Tubulin-Binding 3,5-Bis(styryl)pyrazoles as Lead Compounds for the Treatment of
Castration-Resistant Prostate Cancer
#
#
Vivian W.Y. Liao, Anuradha Kumari, Rajeshwar Narlawar, Soma Vignarajan, David E. Hibbs,
Dulal Panda, and Paul W. Groundwater
Sydney Pharmacy School, Faculty of Medicine and Health, The University of Sydney, Sydney,
NSW 2006, Australia (VWYL, RN, DEH, PWG); Department of Biosciences and Bioengineering,
Indian Institute of Technology Bombay, Mumbai 400076, India (AK, DP); Charles Perkins Centre,
The University of Sydney, Sydney, NSW 2006, Australia (SV)
1

Molecular Pharmacology Fast Forward. Published on April 2, 2020 as DOI: 10.1124/mol.119.118539
This article has not been copyedited and formatted. The final version may differ from this version.
MOL # 118539
Running Title: Novel Tubulin-Binding 3,5-Bis(styryl)pyrazoles
Corresponding authors:
Paul W GROUNDWATER, Sydney Pharmacy School, Faculty of Medicine and Health, The
University of Sydney, Sydney, NSW 2006, Australia; e-mail: paul.groundwater@sydney.edu.au
Dulal PANDA, Department of Biosciences and Bioengineering, Indian Institute of Technology
Bombay, Mumbai 400076, India; e-mail: panda@iitb.ac.in
Text pages: 26
Tables: 5
Figures: 14
References: 39
Abstract: 232 words
Introduction: 347 words
Discussion: 796 words
2

Molecular Pharmacology Fast Forward. Published on April 2, 2020 as DOI: 10.1124/mol.119.118539
This article has not been copyedited and formatted. The final version may differ from this version.
MOL # 118539
Abstract
The microtubule-binding taxanes, docetaxel, and cabazitaxel, are administered intravenously for
the treatment of castration-resistant prostate cancer (CRPC) as the oral administration of these
drugs is largely hampered by their low and highly variable bioavailabilities. Using a simple, rapid,
and environmentally friendly, microwave-assisted protocol, we have synthesized a number of 3,5-
bis(styryl)pyrazoles 2a-l, thus allowing for their screening for anti-proliferative activity in the
androgen-independent PC3 prostate cancer cell line. Surprisingly, two of these structurally simple
3,5-bis(styryl)pyrazoles (2a and 2l) had GI50 values in the low micromolar range in the PC3 cell
line and were thus selected for extensive further biological evaluation (apoptosis and cell cycle
analysis, and effects on tubulin and microtubules). Our findings from these studies show that 3,5-
bis[(1E)-2(2,6-dichlorophenyl)ethenyl]-1H-pyrazole 2l (i) caused significant effects on the cell
cycle in PC3 cells, with the vast majority of treated cells in the G
2
/M phase (89%), (ii) induces cell
0.4 ± 0.1 μM)
death in PC3 cells even after the removal of the compound, (iii) binds to tubulin (K
d
and inhibits tubulin polymerization in vitro, (iv) had no effect upon the polymerization of the
bacterial cell division protein FtsZ (a homolog of tubulin), (v) is competitive with paclitaxel for
binding to tubulin but not with vinblastine, crocin, or colchicine, and (vi) leads to microtubule
depolymerization in PC3 cells. Taken together, these results suggest that 3,5-bis(styryl)pyrazoles
warrant further investigation as lead compounds for the treatment of CRPC.
Significance statement
The taxanes are important components of prostate cancer chemotherapy regimens, but their oral
administration is hampered by very low and highly variable oral bioavailabilities resulting from
their poor absorption, poor solubility, high first-pass metabolism, and efficient efflux by P-
glycoprotein. New chemical entities (NCEs) for the treatment of prostate cancer are thus required
3

Molecular Pharmacology Fast Forward. Published on April 2, 2020 as DOI: 10.1124/mol.119.118539
This article has not been copyedited and formatted. The final version may differ from this version.
MOL # 118539
and we report here the synthesis and investigation of the mechanism of action of some
bis(styryl)pyrazoles, demonstrating their potential as lead compounds for the treatment of prostate
cancer.
Introduction
Globally, prostate cancer is the second most common male cancer in terms of incidence, and fifth
in terms of mortality, with over 300,000 men dying from the disease every year (Todd et al.,
2018). Tumour growth in the early stages of prostate cancer is androgen-dependent, so the typical
treatment for early metastatic prostate cancer is androgen deprivation therapy with a
gonadotropin-releasing hormone analogue such as leuprolide, often in combination with an anti-
androgen such as flutamide (Crawford et al., 1989; Garnick, 1997, Taplin and Balk, 2004). As the
disease progresses, however, molecular and cellular changes occur so that cancer becomes
androgen-independent / hormone refractory (termed castration-resistant prostate cancer [CRPC])
and unresponsive to current hormone therapy (Todd et al., 2018; Feldman and Feldman, 2001).
This form of prostate cancer is aggressive, highly metastatic, is associated with poor prognosis
(mean survival time of 18-24 months) (Abouelfadel and Crawford, 2008) and there is no effective
treatment (Karantanos et al., 2013).
Paclitaxel has been used in the treatment of prostate cancer, and the semi-synthetic taxanes,
docetaxel, and cabazitaxel, are administered intravenously for the treatment of castration-resistant
prostate cancer (CRPC). The oral administration of these drugs is hampered by their low and
highly variable bioavailabilities (Torne et al., 2010), which are due to their poor absorption (a
result of their poor solubilities), efficient efflux by P-glycoprotein (P-gp), which is abundant in
the gastrointestinal tract, and high first-pass metabolism by CYP450s, CYP2C8 and 3A4.
4

Molecular Pharmacology Fast Forward. Published on April 2, 2020 as DOI: 10.1124/mol.119.118539
This article has not been copyedited and formatted. The final version may differ from this version.
MOL # 118539
We report here the discovery of a 3,5-bis(styryl)pyrazole analogue 2l which induces cell death in
PC3 cells (Pululkuri et al., 2005; Schmitt et al., 2014) even after the removal of the compound. The
compound binds to tubulin (K 0.4 ± 0.1 μM), inhibits tubulin polymerization in vitro (with no effect
d
upon the polymerization of the bacterial homolog SpnFtsZ or on the activity of alkaline
phosphatase). Pyrazole 2l is competitive with paclitaxel for binding to tubulin (but not with
vinblastine, crocin, or colchicine) and treatment with pyrazole 2l results in microtubule
depolymerization in PC3 cells, suggesting the further investigation of these 3,5-bis(styryl)pyrazoles
as lead compounds for the treatment of CRPC.
Materials and Methods
General synthetic chemistry procedures
The 3,5-bis(styryl)pyrazoles 2a-l were synthesized from the curcuminoids 1 (Supplemental
Methods) using a CEM Discover SP microwave synthesis system. The spectroscopic and
analytical data for pyrazoles 2a (Amolins et al., 2009; Mayadevi et al., 2012), 2b (Amolins et al.,
2009) and 2k (Luo et al., 2013) were identical to that reported previously. Melting points were
determined on a Stuart Scientific SMP10 apparatus. IR spectra were obtained on a FTIR-8400S
-1
1
13
Shimadzu system, using NaCl plates and values are recorded as wave-numbers (cm ). H and C
NMR spectra were recorded at 400 and 100 MHz, respectively, on a Varian 400-MR magnetic
resonance spectrometer, with chemical shifts (δ) reported in parts per million (ppm). Spectra were
acquired in solutions of deuterated solvents and the residual solvent peaks were used as internal
references; 7.24 ppm (CDCl
3
), 2.50 ppm (DMSO-d
6
), 2.05 ppm (acetone-d ) and 3.31 ppm
6
(
methanol-d ). Low-resolution electrospray ionization (ESI) mass spectrometry was performed on
4
a TSQ Quantum Access Max LCMS/MS (Thomas R Watson Mass Spectrometry Laboratory,
Sydney Pharmacy School, The University of Sydney) and high-resolution mass spectroscopy was
5

Molecular Pharmacology Fast Forward. Published on April 2, 2020 as DOI: 10.1124/mol.119.118539
This article has not been copyedited and formatted. The final version may differ from this version.
MOL # 118539
performed on a Bruker 7T Fourier transform ion cyclotron resonance mass spectrometer (Mass
Spectrometry Unit, School of Chemistry, The University of Sydney). Elemental analysis was
performed on a Thermo Flash 2000 Elemental Analyser (Campbell Microanalytical Laboratory,
University of Otago, New Zealand). All reagents were commercially available and purchased from
Sigma Aldrich (Castle Hill, Sydney, Australia), Alfa Aesar (VWR, Australia). Solvents were
purchased from Chem-Supply (Australia) and were used as received. Flash column
chromatography was performed using Grace Davison (LC604 40-63 µm) Davisil chromatographic
silica media. Thin layer chromatography (TLC) was performed using Grace Davidson Reveleris,
aluminum-backed TLC plates (UV254).
General experimental procedure for the synthesis of pyrazoles 2a-l
Pyrazole derivatives 2a-l were synthesized using a modification of the method of Mayadevi et al.
(2012). Curcuminoids 1 (1.0 mmol) were dissolved in DMF (1 mL). Acetic acid (1 mL) and
hydrazine monohydrate (0.2 mL) were added to the solution in a microwave synthesizer tube. The
mixture was placed in the microwave reactor and heated at 80 ˚C for 5 minutes with rapid stirring.
The reaction was cooled then added dropwise to stirred water (100 mL). The solid was collected
by filtration, purified by flash chromatography using a gradient mobile phase of ethyl acetate and
hexane (1:4 to 1:1) and then recrystallized from ethanol. Spectroscopic and analytical data for the
pyrazoles are provided in the Supplemental Methods.
General experimental procedure for the degradation studies
HiPerSolv® CHROMANORM® LC-MS grade acetonitrile and methanol were purchased from
VWR International. HPLC grade glacial acetic acid was purchased from Fisher Scientific, UK.
Ultrapure water was obtained with a SG, Ultra Clear water system. The chromatographic separation
6

Molecular Pharmacology Fast Forward. Published on April 2, 2020 as DOI: 10.1124/mol.119.118539
This article has not been copyedited and formatted. The final version may differ from this version.
MOL # 118539
was performed on a Aquacil C18 column (2.1 × 100 mm, 5 µm particle size, Thermo Scientific,
USA).
An orbital mixter incubator (Ratek instruments, Australia) was used for the degradation study at
37 ˚C and 200 rpm. The HPLC analysis was carried out on a Shimadzu UFLC prominence system,
consisting of a DGU-20A5R degassing unit, an LC-20AD pump, a SIL-20AHT autosampler, a
SPD-M20A photodiode array (PDA) detector, and a PC with LabSolutions CS software for data
acquisition and processing.
Chromatographic conditions and preparation of standards
The chromatographic method was adapted from Jayaprakasha et al. (2002). The mobile phase
consisted of varying gradients of acetonitrile (A) and 2 % aq. acetic acid solution (B), with a flow
rate of 0.75 mL/min. The total run time was 25 minutes, the last 10 minutes equilibrating the
column for the next run. The wavelength of detection was the λmax of each compound (2a; 325 nm;
2l 300 nm). The injection volume was 10 µL. Table 1 summarizes the chromatographic conditions
used for each of the compounds.
Methanolic standard stock solutions (1 mg/mL) were prepared for each of the analyzed
compounds. The stock solutions were diluted in methanol to obtain concentrations of 0.01, 0.02,
0.03, 0.04, 0.05, and 0.06 mg/mL. For less concentrated standard samples of 2a and 2l, a 1 µg/mL
standard stock was prepared from the 1 mg/mL stock and the stock solutions were diluted in
methanol to obtain 100, 50, 20, 10, 5, and 2 ng/mL solutions. All standards were freshly prepared
from the stock solutions on the day of analysis. The stock solutions were stored at -20 ˚C. The
AUC versus the concentration of each compound standards was plotted and fitted with linear
regression analysis using GraphPad Prism to calculate the slope, intercept, and correlation
coefficient and thus determine the linearity of the standard curve (Supplemental Tables 1 and 2).
7

Molecular Pharmacology Fast Forward. Published on April 2, 2020 as DOI: 10.1124/mol.119.118539
This article has not been copyedited and formatted. The final version may differ from this version.
MOL # 118539
For reproducibility, the experiment was repeated three times, on three separate days, with freshly
prepared standard solutions for each compound.
Degradation study
The degradation study was carried out according to the method outlined by Tønnesen and Karlsen
(
1985). Compounds 2a and 2l were dissolved in methanol (1 mg/mL concentration). The
methanolic solution (100 µL) of the compounds was then added to pre-warmed PBS (900 µL, 37
C) and placed in the orbital incubator. At time points of 1, 2, 4, 8, 12, 24, 48, and 72 hours, 100
˚
µL of the solution was removed and extracted twice with ethyl acetate (2 × 200 µL). The ethyl
acetate extracts were combined, evaporated and the residues were redissolved in methanol (200
µL) for HPLC analysis. 10 µL of the original methanolic solutions were removed, evaporated and
redissolved in methanol (200 µL) and this was taken as t = 0. The area under the curve (AUC) is
the total absorbance of the compound at that given concentration and was used for quantification.
The degradation profiles of each of the compounds are expressed as percentages of compounds
remaining in PBS at the sampled time, with the AUC at t = 0 taken as 100 %.
The detection and quantification limits for analogues 2a and 2l were calculated according to the
method outlined in the guidelines of International Committee on Harmonization (2005) using the
following formulae:
Detection limit = 3.3σ/slope
Quantification limit = 10σ/slope
Where σ is the standard deviation of errors in regression of the standard curve generated upon
linear regression fitting of the average values of three standard curves (n = 3).
8

Molecular Pharmacology Fast Forward. Published on April 2, 2020 as DOI: 10.1124/mol.119.118539
This article has not been copyedited and formatted. The final version may differ from this version.
MOL # 118539
The detection and quantification limits for 2a and 2l were verified by injecting multiple samples
(n = 6) at the concentrations of the detection and quantification limits. The accuracy is expressed
as the percentage of the standard deviation divided by mean values of the six AUC values
determined from the injected samples.
General experimental procedure for the in vitro testing
RPMI-1640, phosphate buffered saline (PBS), trypsin/EDTA 10 × solution, and 3-(4,5-dimethyl-
2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide (MTT) were purchased from Sigma Aldrich
(Castle Hill, Sydney, Australia). Foetal bovine serum (FBS) of Australian origin was purchased
from Bovogen. PC3 cells were purchased from ATCC (Cat. No. CRL-1435). Muse® Count &
Viability Assay Kit, Muse® Annexin V & Dead Cell Assay kit and Muse® Cell Cycle Assay Kit
were purchased from Millipore, Australia. A CLARIOstar® microplate reader was used to
measure the absorbance of MTT formazan at 540 nm. A Muse™ Cell Analyzer (Merck, Millipore)
was used to conduct the Annexin V apoptosis and cell cycle analysis assays.
Anti-proliferative assay
PC3 cells were grown in RPMI-1640 supplemented with 10 % FBS to ~70-80 % confluency before
seeding. Anti-proliferative activity was determined by the MTT colorimetric assay (Mosmann,
1
983). The MTT stock solution (12 mM) was prepared by dissolving 3-(4,5-dimethylthiazol-2-yl)-
,5-diphenyltetrazolium bromide in PBS (5 mg/mL) filtered through a 0.2 µm syringe filter.
2
The pyrazoles and methotrexate were dissolved in DMSO to make a 200 mM stock. PC3 cells were
seeded in 96-well plates at 3,000 cells/well and incubated for 24 hours. The DMSO stock solutions
were diluted in 10 % FBS supplemented RPMI 1640 media; six concentrations of each of the test
compounds were added to wells (in triplicate) with the final DMSO concentration of less than 0.1
9

Molecular Pharmacology Fast Forward. Published on April 2, 2020 as DOI: 10.1124/mol.119.118539
This article has not been copyedited and formatted. The final version may differ from this version.
MOL # 118539
%
in each well. 0.1 % of DMSO in RPMI medium, was also prepared and used as a vehicle control
for each plate and represents 100% growth in this assay. The plates were then incubated for 72
hours (T72).
After 72 hours, the medium was removed from each well and was replaced with 10 % MTT stock
solution in serum and phenol red-free RPMI 1640 medium (100 µL/ well). The plates were then
incubated for 3 hours at 37 ˚C before removing the medium and replacing it with DMSO (50 µL/
well). After mixing, the plates were placed in the microplate reader and the absorbance was
determined. A control plate was prepared in the same way as above and read 24 hours (T ) after
0
seeding. The experiment was repeated at least three times on three separate days to give n = 3.
The percentage of inhibition of cell growth was then calculated using the following formula:
ꢂꢃꢄ (ꢅꢆꢇ)ꢈꢂꢃ (ꢅꢉ)
%
Growth inhibition = 100 − ꢀꢁ
ꢎ × 100ꢏ
ꢂꢃꢄ (ꢊꢋꢌꢍ)ꢈꢂꢃꢄ (ꢅꢉ)
GraphPad Prism was used to determine the GI50 values using a non-linear regression fit after first
transforming concentrations to log scale values. All results were representative of three
independent experiments conducted on three separate days.
Annexin V / propidium iodide apoptosis assay
PC3 cells were seeded in 6-well plates at cell densities of 100,000 cells per well and incubated for
24 hours. The pyrazoles 2a and 2l were added to the wells at their GI50 concentration. 0.1 % of
DMSO in RPMI medium was used as a vehicle control. The cells were incubated for 72 hours, then
they were trypsinized, centrifuged, stained with the Muse™ Annexin V/Dead Cell kit according to
manufacturer’s instructions and then analyzed using, Muse™ Cell Analyzer. The experiment was
10

Molecular Pharmacology Fast Forward. Published on April 2, 2020 as DOI: 10.1124/mol.119.118539
This article has not been copyedited and formatted. The final version may differ from this version.
MOL # 118539
conducted in duplicate, with data acquisition of 4000 events and the experiment was repeated three
times on three separate days to give n = 3.
A set of experiments was performed following the same procedure as above but after 72 hours of
drug treatment, the medium containing the pyrazoles was removed and replaced with fresh medium
and the cells were incubated for a further 72 hours. Cells were then trypsinized, stained with the
Muse™ Annexin V/Dead Cell kit, and analysed using the Muse™ Cell Analyzer. The experiment
was conducted in duplicate, with data acquisition of 4000 events and the experiment was repeated
three times on three separate days to give n = 3.
Cell cycle analysis
PC3 cells were treated in the same way as in the apoptosis assay except, after trypsinization, cells
6
were counted, centrifuged and fixed with cold ethanol solution 70 % (v/v) at 1 × 10 cells/mL. The
˚
cells were left to fix overnight at -20 C, then centrifuged, stained with the Muse™ Cell Cycle kit
according to the manufacturer’s instructions and then analyzed using the Muse™ Cell Analyzer.
The experiment was conducted in duplicate with data acquisition of 5000 events and the experiment
was repeated at least four times on four separate days, to give n = 4.
Experimental procedure for the tubulin and microtubule studies
The determination of the effects of the pyrazoles on the assembly kinetics of tubulin in vitro, the
dissociation constant (K ) for the binding of the analogues to tubulin, and effect on microtubules
d
in cells were performed as described previously (Panda et al., 2005)
Tubulin purification
Tubulin was isolated from goat brain by two cycles of polymerization and depolymerization as
described previously (Panda et al., 2005).
11

Molecular Pharmacology Fast Forward. Published on April 2, 2020 as DOI: 10.1124/mol.119.118539
This article has not been copyedited and formatted. The final version may differ from this version.
MOL # 118539
Light scattering
Tubulin (12 µM) was incubated with different concentrations of 2a and 2l in PEM (50 mM PIPES,
mM EGTA and 3 mM MgCl ) buffer, pH 6.8 on ice for 10 minutes. After 10 minutes of
1
2
incubation, 10% DMSO and 1 mM GTP was added to the reaction mixture and polymerization was
e
monitored at 37 °C in a spectrophotometer, Spectramax M2 , at 350 nm for 30 minutes. The
experiment was repeated three times and the percentage inhibition of tubulin polymerization was
calculated. The IC50 of polymerization was determined by fitting the data in dose-response
inhibition curve in the GraphPad Prism Software.
Sedimentation assay
Tubulin (12 µM) was polymerized at 37 °C for 30 minutes in the presence and the absence of 2a
and 2l as in light scattering. After 30 minutes, the polymerized microtubules were pelleted by
centrifuging it at 33000 g, 30 °C for 30 minutes. The supernatant was separated and the amount of
polymerized microtubules was calculated by loading the supernatant and pellet on the SDS-PAGE
gel followed by intensity quantification of bands by ImageJ or by quantifying the supernatant
protein concentration by Bradford’s assay.
Determination of dissociation constant
Tubulin (2 µM) was incubated with and without different concentrations of 2a or 2l in 25 mM
PIPES buffer pH 6.8 for 10 minutes at room temperature. After incubation, the spectra of
tryptophan emission was recorded in a spectrofluorometer (JASCO FP-6500, Tokyo, Japan) by
exciting the sample at 295 nm and recording the spectra. The fluorescence intensity at λmax was
noted and it was fitted in the following equation in GraphPad Prism software to determine the K .
d
ꢇ
(
[ꢐ ]ꢑ[ꢒ ]ꢑꢓ )ꢈꢕ([ꢐ ]ꢑ[ꢒ ]ꢑꢓ ) ꢈꢖ[ꢐ ][ꢒ ]
ꢉ
ꢉ
ꢔ
ꢉ
ꢉ
ꢔ
ꢉ
ꢉ
ΔF = ∆퐹푚푎푥 ×
ꢗ[ꢐꢉ]
12

Molecular Pharmacology Fast Forward. Published on April 2, 2020 as DOI: 10.1124/mol.119.118539
This article has not been copyedited and formatted. The final version may differ from this version.
MOL # 118539
Where ∆Fmax is the highest fluorescence intensity, P
0
is the concentration of protein and L is the
0
concentration of 2a or 2l.
Determination of tubulin binding site of pyrazole 2l
Competitive inhibition with colchicine
Tubulin (5 µM) was incubated without or with 10 and 20 µM of 2l on ice for 10 min in 25 mM
PIPES buffer pH 6.8 and then, incubated with 10 µM colchicine for 45 min at 37 °C. The
fluorescence spectra (370 – 500 nm) were monitored using a fluorescence spectrophotometer
(JASCO FP-6500, Tokyo, Japan) with excitation of the samples at 340 nm.
Competitive inhibition with vinblastine and crocin
First, we examined whether 2l could inhibit the binding of vinblastine to tubulin using BODIPY-
FL-Vinblastine, a fluorescent analogue of vinblastine (Lin and Chen, 2013). Tubulin (3 µM) was
incubated with 200 nM of BODIPY-FL-vinblastine (Thermo Scientific; cat. no. V12390) on ice
for 10 min in 25 mM PIPES buffer pH 6.8. The reaction mixture was then incubated without or
with 1 and 5 µM of 2l for 10 min on ice. The change in fluorescence of BODIPY-FL-Vinblastine
was observed in a spectrofluorometer (JASCO FP-6500, Tokyo, Japan) by exciting at 488 nm and
taking emission spectra from 498 nm to 540 nm.
Recently, it has been reported that vinblastine and crocin share their binding sites on tubulin
(Sawant et al., 2019; Hire et al.; 2017). Therefore, we examined whether 2l could inhibit the
binding of crocin to tubulin. Crocin (5 µM) was incubated with 10 µM tubulin for 10 min on ice
in PEM buffer pH 6.8. The reaction mixture was then incubated with 10 µM 2l for 10 min on ice
and then change in O.D from 410 nm – 500 nm was observed in a spectrophotometer (JASCO, V-
730).
13

Molecular Pharmacology Fast Forward. Published on April 2, 2020 as DOI: 10.1124/mol.119.118539
This article has not been copyedited and formatted. The final version may differ from this version.
MOL # 118539
Docking of 2l with tubulin
The structure of 2l was drawn in PubChem Sketcher (V2.4) and the PDB coordinates of 2l were
generated in PRODRG server (Schuttelkopf et al., 2004). The Autodock Vina tool was used for
docking of 2l on tubulin (Trott and Olson, 2010) as described previously (Rane et al., 2017, Hura
et al., 2018). The crystal structure of tubulin (PDB ID 5LYJ) was used for docking 2l with tubulin.
The coordinates of all other proteins and molecules (tubulin-tyrosine ligase, CA-4, glycerol,
stathmin, etc.) in 5LYJ, except one  and one  subunit of tubulin were deleted using PyMOL
(DeLano, 2002). The final coordinates used for docking had one subunit, one  subunit, one
GDP, one GTP, one calcium ion and two magnesium ions. Initially global docking was performed
by covering the whole molecule in a grid box of 72 × 102 × 100 Å, with grid spacing of 1 Å. The
global docking was performed 5 times with exhaustiveness of 100 (the number of times the
calculation is repeated) and each set of docking produced 9 conformations. The greatest number of
conformations were found to interact with two sites on the protein (16 conformations at the
interface of the - and -dimers and 19 conformations at the taxol site) (Jaghoori et al., 2016).
These two sites were thus chosen to perform local docking, which was performed in a grid box of
72 × 56 × 58 Å for colchicine and 126 × 86 × 82 Å for epothilone, with a grid spacing of 0.375 Å.
The conformation with minimum binding energy was chosen to analyse the interaction with
tubulin. Control docking with epothilone was performed in a similar way to validate the docking
protocol. The coordinates of epothilone and colchicine were obtained from their PDB structures
4O4I and 1SA0, respectively. The interaction of the docked conformers with tubulin was further
analyzed in UCSF Chimera version 1.11 (Pettersen et al., 2004) to determine the amino acid
residues in the binding pocket and the residues within 4 Å of 2l to determine the possible hydrogen
bonding interactions.
14

Molecular Pharmacology Fast Forward. Published on April 2, 2020 as DOI: 10.1124/mol.119.118539
This article has not been copyedited and formatted. The final version may differ from this version.
MOL # 118539
Competitive inhibition with paclitaxel
Tubulin (12 µM) was incubated with different concentrations (0, 2, 5, 10, 20 and 30 µM) of 2l for
0 min on ice in PEM buffer pH 6.8. Different concentrations (2, 5 and 7 µM) of paclitaxel was
1
added to the reaction mixtures for each concentration of 2l and incubated for 10 minutes on ice.
The polymerization reaction was monitored immediately after adding 1 mM GTP in Spectramax
M2e by taking O.D at 350 nm. The percentage inhibition of tubulin polymerization versus
concentration of 2l was plotted in a Lineweaver-Burk plot at different concentrations of paclitaxel
to determine the nature of inhibition.
Effects of 2l on the polymerization of Streptococcus pneumoniae FtsZ
Purified FtsZ (10 µM) was polymerized in the presence of 1 mM GTP at 37 °C as described
previously (Dhaked et al, 2019). The polymerization reaction was monitored in a
spectrofluorometer (JASCO FP-6500, Tokyo, Japan) with the wavelength for both excitation and
the emission at 400 nm. Polymerization of SpnFtsz was also observed with prior incubation of the
protein with 20 µM 2a and 2l in ice for 10 minutes.
Alkaline phosphatase assay
2
U of alkaline phosphatase was taken in glycine-NaOH buffer pH 10.4 with 1 mM MgCl
2
and 0.1
mM ZnCl . 250 µM of para-nitrophenylphosphate (PNPP) was mixed with the reaction mixture in
2
a cuvette. The reaction mixture was immediately put in a spectrophotometer and the conversion of
the yellow coloured product was monitored by taking the O.D at 410 nm. A similar experiment
was performed with both 2a and 2l by incubating 20 µM each of 2a and 2l for 10 minutes with
alkaline phosphatase on ice.
Sulphorhodamine B (SRB) assay
15

Molecular Pharmacology Fast Forward. Published on April 2, 2020 as DOI: 10.1124/mol.119.118539
This article has not been copyedited and formatted. The final version may differ from this version.
MOL # 118539
The GI50 in HeLa cells was determined by SRB assay as described previously (Vichai and
Kirtikara, 2006). Briefly, HeLa cells were seeded in TC treated 96-well plate (10,000 cells/well).
After attachment, the cells were treated with different concentrations of 2a and 2l and incubated
for 24 hours followed by fixation with 50% trichloroacetic acid for 1 hour at 4 °C. The plates were
then washed, dried and stained with 0.4% SRB for 1 hour at room temperature. After drying of the
plates, the dye was dissolved in 10 mM Tris pH 10.5 and the absorbance was measured at 520 nm
e
in Spectramax M2 . The percentage inhibition of cell proliferation at each concentration was
determined by the formula,
푁 − 푁
ꢚ
ꢛ
%
퐼푛ℎ푖푏푖푡푖표푛 표푓 푐푒푙푙 푝푟표푙푖푓푒푟푎푡푖표푛 = 100 − ꢘꢙ
ꢝ × 100ꢞ
푁 − 푁
ꢜ
ꢛ
Where N
t
is O.D of the wells where 2a or 2l was added. N
0
is the O.D of the wells at the time of
2a or 2l addition, and N
c
is the O.D of the wells of 0.1% DMSO treatment. This percentage
inhibition was then fitted in dose-response inhibition curve in GraphPad Prism software.
Immunofluorescence microscopy
The coverslips were first coated with poly-L-lysine and then cells were added onto coverslips
(25000 cells/well) for attachment. After attachment, cells were treated with different concentrations
of 2a and 2l and incubated for 24 hours. Cells were then fixed with 4% formaldehyde and
permeabilized with 100% chilled methanol. 2% BSA was used for blocking the non-specific
binding of the antibody and then the anti-α-tubulin antibody (Sigma) was added to cells and
incubated for 3 hours at 37 °C. After washing of primary antibody secondary antibody tagged with
FITC (Sigma) was added to the cells and incubated for 1.5 hours. The coverslips were mounted on
clean glass slides using the mounting media having DAPI, purchased from vector laboratories, and
observed under a fluorescence microscope (Yokogawa CSU-XI).
16

Molecular Pharmacology Fast Forward. Published on April 2, 2020 as DOI: 10.1124/mol.119.118539
This article has not been copyedited and formatted. The final version may differ from this version.
MOL # 118539
Live-dead assay
The percentage of the live and dead cell was determined by PI staining. The cells were seeded up
to 50% confluency in T25 flasks and then incubated with 4 and 8 µM of 2a and 2l for 24 hours.
The cells were trypsinized, pelleted and resuspended in PBS with a final concentration of cells
approximately 1 million per mL. The cells were stained with PI and then analyzed by flow
cytometry for PI positive and PI negative cells. 10,000 cells were counted in each case. The
experiment was repeated three times.
Statistical analysis
‘±’ represents standard deviation (S.D.) and p values were calculated using Student’s t-test.
Results
Synthetic chemistry
The synthesis of the bis(styryl)pyrazoles 2a-l takes advantage of our recently described
microwave-assisted synthesis of curcuminoids (Groundwater et al., 2017), in which the total
reaction time was shortened to 10 minutes and the analytically pure compounds were mostly
obtained by the simple recrystallization from ethanol of the crude solid obtained from the acid
hydrolysis of the boron-containing intermediate. The ready availability of these synthetic
intermediates 1a-l lead to the preparation of non-curcuminoid analogues such as 3,5-
bis(stryryl)pyrazole analogues 2a-l for biological evaluation, Scheme 1, via the cyclization with
hydrazine in an acid-catalyzed microwave-assisted condensation, to generate the respective 2a-l
(Scheme 1).
Anti-proliferative activity of 3,5-bis(styryl)pyrazoles 2a-l in PC3 cells
17

Molecular Pharmacology Fast Forward. Published on April 2, 2020 as DOI: 10.1124/mol.119.118539
This article has not been copyedited and formatted. The final version may differ from this version.
MOL # 118539
The growth inhibitory effect of all the 3,5-bis(styryl)pyrazoles 2a-l on the PC3 cell line was
determined by the MTT assay (Table 2) which was validated by the trypan blue exclusion cell
count. Methotrexate was used as a positive control, with a GI50 value (obtained in this study) of
0.012 ± 0.008 μM, which is within the ranges of the reported GI50 values in PC3 cells for
methotrexate (0.001 – 0.1 μM) (Serova et al., 2010; Derenne et al., 2010; Marques et al., 2010).
Two compounds (2a and 2l) had GI50 values in the low micromolar range (< 2.5 μM)
(Supplemental Figure 1) and were thus selected for further biological evaluation, as well as
degradation and solubility studies.
Degradation studies
We evaluated the stability and solubility of the two analogues selected for extensive biological
testing, 2a and 2l, under simulated physiological conditions (pH 7.4, 37 ˚C), and the degradation
kinetics are summarized in Figure 1. After 24 hours under the simulated physiological conditions,
10 % of analogue 2a remained, while 71 % of analogue 2l remained after 72 hours (Supplemental
Figures 2 and 3).
The effect of 3,5-bis(styryl)pyrazoles 2a and 2l on the induction of apoptosis in PC3 cells
The two selected analogues (2a and 2l) did not induce apoptosis at their GI50 concentrations, as
determined by the annexin V / propidium iodide (PI) apoptosis assay. PC3 cells were treated at the
GI50 of analogues 2a and 2l, with cells treated with DMSO (0.01 %) as the vehicle control. The
cells were then incubated for 72 hours, stained with annexin V/PI, and analyzed. The representative
dot plots depicting the results from this apoptosis assay are shown in Figure 2; cells treated with
and without these compounds maintained more than 90 % viability, indicating that they did not
undergo apoptosis when treated at the GI50 concentrations. However, the cells died after treatment
with higher concentrations of 2a and 2l (4 and 8 μM), as shown by the increasing number of PI
18

Molecular Pharmacology Fast Forward. Published on April 2, 2020 as DOI: 10.1124/mol.119.118539
This article has not been copyedited and formatted. The final version may differ from this version.
MOL # 118539
positive cells after PI staining and counting 10,000 cells by flow cytometry (Figure 3, Table 4). In
addition, we quantified the extent of apoptosis in three other cell lines, namely lung carcinoma
(
A549), skin melanoma (B16F10) and liver carcinoma (Huh-7) at similar concentrations of 2a and
l. The compounds induced apoptosis in all three cell lines in a concentration dependent manner
Supplemental Figure 4 and Supplemental Table 3).
2
(
The effect of 3,5-bis(styryl)pyrazoles 2a and 2l on the cell cycle of PC3 cells
A cell cycle analysis was conducted in order to determine the effects of 3,5-bis(styryl)pyrazoles
2
a and 2l on cell division. PC3 cells were treated for 72 hours at the GI50s of these compounds,
with cells treated with 0.01 % of DMSO used as the vehicle control. The cell cycle analysis showed
a significant increase in the percentage of cells in the G /M phase when treated with the 3,5-
2
bis(styryl)pyrazoles at their GI50 concentrations, in comparison to the control (Figure 4;
Supplemental Table 4).
Assessment of the outcome of mitotic arrest induced by 3,5-bis(styryl)pyrazoles 2a and 2l.
In order to evaluate the potential of 3,5-bis(styryl)pyrazoles 2a and 2l as effective agents for the
treatment of CRPC, we examined whether cell death results from mitotic arrest in PC3 cells. PC3
cells were treated for 72 hours with 3,5-bis(styryl)pyrazoles 2a and 2l (at their GI50s) and the
medium containing the compounds was then removed and replaced with fresh medium. The cells
were incubated for a further 72 hours before being stained with annexin V / PI and analysed. The
percentage of cell death 72 hours after the removal of 2a and 2l (Table 3 and Supplemental Figure
5
4
) was lower than the mitotic block after 72 hours treatment with 2a and 2l (Supplemental Table
), suggesting that the effects of the compounds on PC3 cells are reversible.
19

Molecular Pharmacology Fast Forward. Published on April 2, 2020 as DOI: 10.1124/mol.119.118539
This article has not been copyedited and formatted. The final version may differ from this version.
MOL # 118539
Determination of the effect of 3,5-bis(styryl)pyrazoles 2a and 2l on tubulin polymerization.
Since both the compounds, 2a and 2l, caused a mitotic block, a characteristic of most anti-tubulin
agents (Mukhtar et al., 2014), the effect of 2a and 2l on tubulin was investigated. 3,5-
Bis(styryl)pyrazole 2l inhibited the polymerization of purified tubulin (Figure 5), inhibiting both
the rate and extent of tubulin assembly, as monitored by light scattering. The extent of inhibition
of tubulin polymerization by the compound was also determined by the sedimentation assay
(Supplemental Figure 6). The half maximal inhibitory concentration (IC50) of 3,5-
bis(styryl)pyrazole 2l for tubulin polymerization was found to be 4.7 ± 1.2 μM by light scattering
and 4 ± 1.5 μM by the sedimentation assay.
The effect of analogue 2a on the polymerization of purified tubulin was also determined by light
scattering and sedimentation assay (Supplemental Figure 7). Unlike the pyrazole 2l, analogue 2a
was not able to inhibit the tubulin polymerization in vitro (Supplemental Figure 7).
Effect of 3,5-bis(styryl)pyrazoles 2a and 2l on microtubules of PC3 cells
Next, the effects of 3,5-bis(styryl)pyrazoles 2a and 2l on the microtubules of PC3 cells were
examined by fluorescence microscopy. Cells were seeded on poly-L-lysine coated coverslips and
treated with 4 and 8 μM of 2a or 2l. Both the compounds caused the depolymerization of the
interphase microtubules in PC3 cells (Figure 6a) as in the control cells, thread-like microtubule
structures can be seen which were well spread in the cytoplasm of the cells. These thread-like
structures were not properly visible in cells treated with 2a and 2l, indicating that the interphase
microtubules were depolymerized by the treatment with these drugs (Figure 6a). Although both 2a
and 2l also strongly depolymerized spindle microtubules and caused the formation of abnormal
spindles in PC3 cells (Figure 6b), the effects of 2l on interphase microtubules were more
pronounced (Supplemental Figure 8a). A normal mitotic cell with spindle microtubules is shown
20

Molecular Pharmacology Fast Forward. Published on April 2, 2020 as DOI: 10.1124/mol.119.118539
This article has not been copyedited and formatted. The final version may differ from this version.
MOL # 118539
in the Figure 6b control panel (arrow). In treated cells, abnormal multipolar spindles can be seen
and the number of mitotic cells per field increased significantly after the treatment with 2a and 2l
which also confirms that these compounds cause mitotic block (Figure 6b, Supplemental Figure
8b). In the control cells, DNA is aligned on the metaphase plate (arranged more compactly in a
straight line [arrow]) while after treatment the DNA is scattered and unable to align. We also
determined the effect of 2a and 2l on the microtubules of HeLa cells. The GI50 of 2a and 2l was
determined to be 2.6 ± 1.2 and 2.4 ± 0.5 µM, respectively (Supplemental Figure 9 a-b). Both 2a
and 2l depolymerized interphase and spindle microtubules in HeLa cells (Supplemental Figure 9
c-d), with the depolymerization of microtubules by 2l being more pronounced than that produced
by 2a (Supplemental Figure 10a). The population of mitotic cells also increased upon treatment
with 2a and 2l, suggesting that the compounds cause mitotic block (Supplemental Figure 10b).
Kinetics of microtubule depolymerization and cell death after treatment of PC3 cells with
3,5-bis(styryl)pyrazole 2a
To examine whether or not microtubules are the primary targets of 2a and 2l, we compared the
kinetics of microtubule disassembly and cell death upon treatment with both compounds. PC3 cells
were incubated without or with either 4 and 8 µM of 2a and 2l for 2 and 4 hours. The fluorescence
intensity of microtubules was found to be strongly reduced after both 2 and 4 hour treatment with
2
a and 2l (Figure 7, Supplemental Figure 11). The fluorescence intensity of microtubules in
vehicle treated cells was determined to be 1525 ± 229 (a.u.), dropping to 1054 ± 112 and 913 ±
05 (a.u.) after treatment with 4 µM 2a for 2 and 4 hours, respectively, and 825 ± 61 and 771 ± 54
a.u.) after treatment with 4 µM 2l for 2 and 4 hours, respectively (Figure 7, Supplemental Figure
1
(
1
8
1). The fluorescence intensity of the microtubules was reduced by 54 and 57% in the presence of
µM 2a for 2 and 4 hours, and 41% and 72% in the presence of 8 µM 2l for 2 and 4 hours,
21

Molecular Pharmacology Fast Forward. Published on April 2, 2020 as DOI: 10.1124/mol.119.118539
This article has not been copyedited and formatted. The final version may differ from this version.
MOL # 118539
respectively (Figure 7, Supplemental Figure 11). These results indicate that treatment with both
2a and 2l led to the formation of depolymerized microtubules within a short time period. In
contrast, treatment with 4 and 8 µM 2a and 2l, for 2 or 4 hours, did not induce cell death in PC3
cells (Figure 8 and Supplemental Table 5), indicating that the depolymerization of microtubules
preceded cell death upon treatment with 2a or 2l, and that microtubules are the primary targets of
both 2a and 2l in PC3 cells.
Binding of 3,5-bis(styryl)pyrazole 2l to tubulin
Since 2a and 2l caused microtubule depolymerization, we characterized the binding of 2a and 2l
with tubulin; the binding of 2l with tubulin was examined by monitoring the tryptophan
fluorescence of tubulin (Bhattacharyya et al., 2010). The fluorescence of tubulin was found to be
reduced in the presence of 2l indicating that the compound binds to tubulin (Figure 9). The change
in the fluorescence intensity of tubulin in the presence of 2l was plotted in a binding isotherm,
which yielded a dissociation constant (K
The binding affinity of 2a to tubulin was also determined (Supplemental Figure 12) and it was
found that it binds to tubulin with a much weaker affinity (K , 4.6 ± 1.1 μM) than 2l (K , 0.4 ± 0.1
d
) of 0.4 ± 0.1 μM for the binding interaction.
d
d
μM), a possible explanation for the weak effect of 2a on the polymerization of purified tubulin and
interphase cellular microtubules.
Determination of binding site of 3,5-bis(styryl)pyrazole 2l on tubulin
Pyrazole 2l binds to tubulin with higher affinity than 2a and we have determined the binding site
of 2l on tubulin. Tubulin has three well-defined binding sites for small molecules; the taxane, vinca
alkaloid, and colchicine sites (Supplemental Figure 13). Compounds that bind to the vinca alkaloid
22

Molecular Pharmacology Fast Forward. Published on April 2, 2020 as DOI: 10.1124/mol.119.118539
This article has not been copyedited and formatted. The final version may differ from this version.
MOL # 118539
and colchicine sites generally destabilize microtubules so we first checked the binding of 2l to
these binding sites.
The fluorescence intensity of colchicine and BODIPY-FL-vinblastine increases upon binding to
tubulin, while the absorbance of crocin, a vinca domain binder, increases upon binding to tubulin.
If 2l inhibits the binding of colchicine to tubulin, the fluorescence of colchicine-tubulin complex
should be reduced in the presence of 2l. However, the prior incubation of 2l with tubulin did not
reduce the fluorescence of colchicine-tubulin complex, indicating that 2l does not bind to the
colchicine site on tubulin (Supplemental Figure 14a). To check the binding of 2l at the vinblastine
site, we used BODIPY-FL-vinblastine and crocin. There was no change in the fluorescence of the
BODIPY-FL-vinblastine-tubulin complex or the absorbance of the crocin-tubulin complex,
respectively, upon prior incubation with 2l, indicating that 2l does not bind to the vinblastine site
on tubulin (Supplemental Figure 14b,c).
Consequently, in order to determine the putative binding site, we performed molecular docking
with the tubulin crystal structure (PDB ID 5LYJ) as described previously (Rane et al., 2017, Hura
et al., 2018). We found that most conformations of 2l were interacting with two distinct sites on
tubulin; the paclitaxel binding site on -tubulin (Figure 10) and the interface of the α- and β-tubulin
heterodimer (Supplemental Figure 15), near the colchicine site. To further validate the docking,
we docked epothilone, a paclitaxel site binder, with tubulin. The RMSD of the docked
conformation of epothilone to the epothilone conformation present in complex with tubulin (PDB
ID 404I) was found to be 1 Å, confirming that the docking protocols were appropriate (Figure 11)
(Hura et al., 2018). Since competitive binding data with colchicine showed that 2l does not bind
to the colchicine site (Supplemental Figure 14a), we further analysed the docking data only for
paclitaxel site.
23

Molecular Pharmacology Fast Forward. Published on April 2, 2020 as DOI: 10.1124/mol.119.118539
This article has not been copyedited and formatted. The final version may differ from this version.
MOL # 118539
We found that 2l possibly forms 2 hydrogen bonds in the paclitaxel binding site, with the peptide
backbone of T273 (2.2 Å) and P271 (2.6 Å) (Figure 12). Further analysis of the docking gave the
binding energy of 2l with tubulin, which was found to be -8.4 kcal/mol, while that of epothilone
was found to be -9.1 kcal/mol. Taxol also forms a similar hydrogen bond with the peptide
backbone of T273 (3.0 Å). In addition, paclitaxel also forms other hydrogen bonds, with R358 (3.3
Å), Q278 (3.4 Å), H226 (2.6 Å) and the peptide backbone of R358 (3.1 Å) (Snyder et al., 2001;
Yadava et al., 2015).
We also analyzed the binding pocket of both epothilone and 2l on beta-tubulin and found that there
were 12 residues which are common to both 2l and epothilone (Table 5) showing that 2l and
epothilone share the same binding pocket. Since the taxol binding pocket has hydrophobic residues
in the H7 helix, M loop and β strand of S7, S9-10 (Ranade et al., 2016), it is possible that 2l also
makes some hydrophobic interactions within the binding pocket.
To elucidate the binding site of 2l on tubulin, we performed a competitive inhibition experiment
of 2l with paclitaxel. The effects of 2l on the polymerization of tubulin in the presence of different
concentrations of paclitaxel were determined. 2l exerted a stronger inhibitory effect on tubulin
polymerization in the presence of low concentrations of paclitaxel. The percentage inhibition at
each concentration of taxol was plotted in the Michaelis–Menten equation (Figure 13a). The
resulting curves showed a typical competitive inhibition pattern, where Vmax (maximum
percentage of inhibition) remains unchanged (p > 0.05). Furthermore, the Lineweaver-Burk plot
indicated that the nature of inhibition is competitive (Figure 13b) as the plot shows that the affinity
of the substrate 2l changes with changes in the paclitaxel concentration, as suggested by a decrease
in the x-intercept with increasing concentration of paclitaxel.
Determination of the effect of 3,5-bis(styryl)pyrazoles 2a and 2l on FtsZ
24

Molecular Pharmacology Fast Forward. Published on April 2, 2020 as DOI: 10.1124/mol.119.118539
This article has not been copyedited and formatted. The final version may differ from this version.
MOL # 118539
3,5-Bis(styryl)pyrazoles, 2a and 2l, neither inhibited the polymerization of FtsZ (a bacterial
homolog of tubulin) nor the enzymatic activity of alkaline phosphatase (Figure 14) suggesting that
the compounds are not PAINS.
Discussion
A range of 3,5-bis(styryl)pyrazole analogues 2a-l have been prepared using a very simple
environmentally-friendly microwave-assisted protocol, allowing the rapid and clean generation of
analogues in minutes without the need for multiple chromatographic steps for purification. All
analogues were screened for anti-proliferative activity in the PC3 (androgen-independent) prostate
cancer cell line and two analogues (2a and 2l) with GI50 values in the low micromolar range were
selected for further biological evaluation. 3,5-Bis(styryl)pyrazole 2a was the most potent
compound in the study of the anti-proliferative activity in PC3 cells (GI50 0.85 ± 0.34 μM), while
the novel analogue 2l was also selected for further biological evaluation (Supplemental Figure 1);
it has been shown in other studies that analogue 2a has GI50 values of 5.6 μM (in PC3 cells) (Fuchs
et al., 2009), 4.19 μM and 0.25 μM in breast cancer cell lines, MCF-7 and SKBR3, respectively
(Amolins et al., 2009). The GI50 of 2a and 2l against human cervical cancer (HeLa) cells was
determined to be 2.6 ± 1.2 and 2.4 ± 0.5 µM, respectively (Supplemental Figure 9).
In a stability study, we found that after 72 hours only 1 % of derivative 2a remained, in comparison
to the 71% of analogue 2l which remained. The instability of 2a at pH 7.4 may be due to the
presence of the para-hydroxy group; a detailed study of the mechanism of degradation of 1a under
physiological conditions suggested that the major degradation pathway is auto-oxidation due to
the presence of this group (Gordon et al., 2015). As 2a is a derivative of 1a, it too could undergo
similar auto-oxidation, leading to rapid degradation. Although the GI50 of 2a is less than that of
2l, 2l might thus be a better lead than 2a. Neither of these analogues induced apoptosis at their
25

Molecular Pharmacology Fast Forward. Published on April 2, 2020 as DOI: 10.1124/mol.119.118539
This article has not been copyedited and formatted. The final version may differ from this version.
MOL # 118539
GI50 concentration but, as can be seen from the cell cycle histograms in Figure 4, both caused
significant effects on the cell cycle, with the vast majority of treated cells in the G
and 89.3 %, respectively), suggesting that these analogues inhibit PC3 cell growth as a result of
/M arrest. With no cell division, as a result of G /M arrest, the percentage of cells in the G /G
phase decreases and this, in turn, reduces the number of cells in the S phase.
2
/M phase (89.6
G
2
2
0
1
We next examined whether cell death results from mitotic arrest in PC3 cells. The results showed
that the tested compounds caused greater cell death in PC3 cells than the control (Table 3;
Supplemental Figure 5). Analogue 2a had the least effect on PC3 cells 72 hours after removal,
despite showing G /M arrest in the cell cycle analysis, indicating that cells treated with this
2
analogue may either have divided or remained in a senescing state. Treatment with analogue 2l led
to the greatest total percentage of dead PC3 cells. Analysis of the cell cycle in cells treated with
analogue 2l showed that this compound resulted in the greatest percentage of cells in the G /M
2
phase, indicating that the arrested cells undergo significant mitotic death.
There are three possible main outcomes from drug-mediated mitotic arrest upon drug removal
Yamada and Gorbsky, 2006), the cells divide (they may recover, leading to normal cell division
(
or divide with abnormalities in the genome), the cells may remain in a senescent state, in which
they are metabolically active but do not undergo further cell division, or cell death may be
triggered, either by necrosis or apoptosis.
3,5-Bis(styryl)pyrazole 2l inhibited the polymerization of purified tubulin (Figure 5), inhibiting
both the rate and extent of tubulin assembly, as monitored by light scattering and sedimentation
assay. Extensive testing of the effects of these pyrazoles on tubulin and microtubules showed that
2l binds to tubulin tightly with a K
d
of 0.4 ± 0.1 μM and inhibits the assembly of purified tubulin
with no effect on the assembly of FtsZ, a bacterial homologue of tubulin which shows the specific
26

Molecular Pharmacology Fast Forward. Published on April 2, 2020 as DOI: 10.1124/mol.119.118539
This article has not been copyedited and formatted. The final version may differ from this version.
MOL # 118539
binding of 2l to tubulin only. Treatment with both pyrazoles 2a and 2l caused depolymerization of
the microtubules in PC3 and HeLa cells. We also found that, although 2l depolymerizes
microtubules, it shares its binding site on tubulin with paclitaxel. Paclitaxel is a known microtubule
polymerizing agent and paclitaxel site binders generally stabilize microtubules, however, few
tubulin binding small molecules such as estramustine (Laing et al., 1997) and griseofulvin
(Rathinasamy et al., 2010) are reported to share their binding sites with paclitaxel but to destabilize
microtubules. In summary, we have identified a potential lead which could be the basis for future
optimization in the search for novel agents for the treatment of CRPC. Both pyrazoles exert similar
effects on HeLa cells (Supplemental Figure 9), GI50s of 2.6 ± 1.2 μM (2a) and 2.4 ± 0.5 μM (2l),
through their interactions with tubulin and the depolymerization of microtubules in cells.
Acknowledgements
We thank Drs Donna Lai and Sheng Hua at the Molecular Biology Facility, Bosch Institute for
their technical support. We also thank CRNTS, IIT Bombay for kindly providing a spinning disc
confocal microscope and FACS facility.
Authorship contributions
Participated in research design: Vivian W.Y. Liao, Anuradha Kumari, Rajeshwar Narlawar, Soma
Vignarajan, David E. Hibbs, Dulal Panda, Paul W. Groundwater
Conducted experiments: Vivian W.Y. Liao, Anuradha Kumari, Rajeshwar Narlawar
Contributed new reagents or tools: Vivian W.Y. Liao, Anuradha Kumari, Rajeshwar Narlawar
Performed data analysis: Vivian W.Y. Liao, Anuradha Kumari, Soma Vignarajan, David E. Hibbs,
Dulal Panda, Paul W. Groundwater
27

Molecular Pharmacology Fast Forward. Published on April 2, 2020 as DOI: 10.1124/mol.119.118539
This article has not been copyedited and formatted. The final version may differ from this version.
MOL # 118539
Wrote or contributed to the writing of the manuscript: Vivian W.Y. Liao, Anuradha Kumari, David
E. Hibbs, Dulal Panda, Paul W. Groundwater
References
Abouelfadel, Z and Crawford, ED (2008) Leuprorelin depot injection: patient considerations in the
management of prostatic cancer. Ther. Clin. Risk Manag. 4:513.
Amolins, MW, Peterson, LB and Blagg, BSJ (2009) Synthesis and evaluation of electron-rich
curcumin analogues. Bioorg. Med. Chem. 17:360-367.
Bhattacharyya, B, Kapoor, S and Panda, D (2010) Fluorescence Spectroscopic Methods to
Analyze Drug-Tubulin Interactions. In Microtubules, in Vitro: Microtubules, in Vitro
(Wilson, L and Correia, JJ, eds.). pp. 301-329, Elsevier Academic Press Inc, San Diego
Crawford, ED, Eisenberger, MA, McLeod, DG, Spaulding, JT, Benson, R, Dorr, FA, Blumenstein,
BA, Davis, MA and Goodman, PJ (1989) A controlled trial of leuprolide with and without
flutamide in prostatic carcinoma. N. Engl. J. Med. 321: 419-424
DeLano, WL, (2002) The PyMOL molecular graphics system, Version 1.1; Schrodinger LLC
Derenne, A, Gasper, R and Goormaghtigh, E (2010) Monitoring of metabolism perturbation in
prostate PC-3 cancer cells by sub-lethal concentrations of methotrexate. J. Spectrosc. 24:
55-60
Dhaked, HPS, Ray, Shashikant, R, Battaje, RR, Banerjee, A and Panda. D (2019) Regulation of
Streptococcus pneumoniae FtsZ assembly by divalent cations: paradoxical effects of Ca2+
on the nucleation and bundling of FtsZ polymers. FEBS Journal, 14928
Feldman, BJ and Feldman, D (2001) The development of androgen-independent prostate cancer.
Nature Rev. Cancer 1: 34-45.
Fuchs, JR, Pandit, B, Bhasin, D, Etter, JP, Regan, N, Abdelhamid, D, Li, C, Lin, J and Li, P-K
(2009) Structure–activity relationship studies of curcumin analogues. Bioorg. Med. Chem.
Lett. 19: 2065-2069.
Garnick, MB (1997) Hormonal therapy in the management of prostate cancer: from Huggins to
the present. Urology 49: 5-15.
28

Molecular Pharmacology Fast Forward. Published on April 2, 2020 as DOI: 10.1124/mol.119.118539
This article has not been copyedited and formatted. The final version may differ from this version.
MOL # 118539
Gordon, ON, Luis, PB, Sintim, HO and Schneider, C (2015). Unraveling curcumin degradation:
autoxidation proceeds through spiroepoxide and vinylether intermediates en route to the
main bicyclopentadione. J. Biol. Chem. 290: 4817–4828.
Groundwater, PW, Narlawar, R, Liao, VWY, Bhattacharya, A, Srivastava, S, Kunal, K,
Doddareddy, M, Oza, PM, Mamidi, R, Marrs, ECL, Perry, JD, Hibbs, DE and Panda, D
(2017) A Carbocyclic Curcumin Inhibits Proliferation of Gram-Positive Bacteria by
Targeting FtsZ. Biochemistry 56: 514-524
Hire, RR, Srivastava, S, Davis, MB, Konreddy, AK and Panda, D (2017) Antiproliferative activity
of crocin involves trageting of microtubules in breast cancer cells. Sci. Rep. 7: 44984.
Hura, N, Naaz, A, Prassanawar, SS, Guchhait, SK, Panda D (2018) Drug clinical agent molecular
hybrdid: Synthesis of diaryl(trifluoromethyl)pyrazoles as tubulin targeting anticancer
agents. ACS Omega. 3: 1955-1969
ICH Harmonised Tripartite Guideline Validation of Analytical Procedures: Text and Methodology
Q2
(R1)
(2005)
International
Conference
on
Harmonization
(https://www.ich.org/fileadmin/Public_Web_Site/ICH_Products/Guidelines/Quality/Q2_
R1/Step4/Q2_R1__Guideline.pdf)
Jaghoori, MM, Bleijlevens and Olabarriaga, SD (2016) 1001 ways to run AutoDock Vina for
virtual screening. J. Computer-Aided Mol Design. 30: 237-249
Jayaprakasha, GK, Jagan Mohan Rao, L and Sakariah, KK (2002) Improved HPLC Method for
the Determination of Curcumin, Demethoxycurcumin, and Bisdemethoxycurcumin. J.
Agric. Food Chem. 50: 3668-3672.
Karantanos, T, Corn, P and Thompson, T (2013) Prostate cancer progression after androgen
deprivation therapy: mechanisms of castrate resistance and novel therapeutic approaches.
Oncogene 32: 5501-5511.
Lin, W and Chen, T (2013) A vinblastinefluorescent probe for pregnane X receptor in a time-
resolved fluorescence resonance enegry transfer assay, Anal. Biochem. 443: 252-260.
Luo, JX, Ding, W, Zhang, YQ, Yang, ZG, Li, Y and Ding, LJ (2013) Semisynthesis and acaricidal
activities of isoxazole and pyrazole derivatives of
a
natural product
bisdemethoxycurcumin. J. Pestic. Sci. 38: 214-219.
Marques, SM, Enyedy, ÉA, Supuran, CT, Krupenko, NI, Krupenko, SA and Amélia Santos, M
2010) Pteridine–sulfonamide conjugates as dual inhibitors of carbonic anhydrases and
(
29

Molecular Pharmacology Fast Forward. Published on April 2, 2020 as DOI: 10.1124/mol.119.118539
This article has not been copyedited and formatted. The final version may differ from this version.
MOL # 118539
dihydrofolate reductase with potential antitumor activity. Bioorg. Med. Chem. 18: 5081-
5089.
Mayadevi, M, Sherin, DR, Keerthi, VS, Rajasekharan, KN and Omkumar, RV (2012) Curcumin
is an inhibitor of calcium/calmodulin dependent protein kinase II. Bioorg. Med. Chem. 20:
6040-6047.
Mosmann, T (1983) Rapid colorimetric assay for cellular growth and survival: Application to
proliferation and cytotoxicity assays. J. Immunol. Methods 65: 55-63.
Mukhtar, E, Adhami, VM and Mukhtar, H (2014) Targeting Microtubules by Natural Agents for
Cancer Therapy. Mol. Cancer Ther. 13: 275-284.
Panda, D, Rathinasamy, K, Santra, MK and Wilson, L (2005) Kinetic suppression of microtubule
dynamic instability by griseofulvin: Implications for its possible use in the treatment of
cancer. Proc. Natl. Acad. Sci. U.S.A. 102: 9878-9883.
Pettersen, EF, Goddard, TD, Huang, CC, Couch, GS, Greenblatt, DM Meng, EC and Ferrin TE
(2004) UCSF Chimera – a visualization system for exploratory research and analysis. J.
Comput. Chem. 25: 1605-1612.
Pulukuri, SM, Gondi, CS, Lakka, SS, Jutla, A, Estes, N, Gujrati, M and Rao, JS (2005) RNA
interference-directed knockdown of urokinase plasminogen activator and urokinase
plasminogen activator receptor inhibits prostate cancer cell invasion, survival, and
tumorigenicity in vivo. J. Biol. Chem. 280: 36529-36540.
Ranade, AR, Higgins, LA, Markowski, TW, Glaser, N, Kashin, D, Bai, R, Hong, KH, Hamel, E,
Hofle, G and Georg, GI (2016) Characterizing the epothilone binding site on β-tubulin by
photoaffinity labeling: Identification of β-tubulin peptides TARGSQQY and TSRGSQQY
as targets of an epothilone photoprobe for polymrized tubulin. J. Med. Chem. 59: 3499-
3514
Rane, JS, Bhaumik, P, Panda, D (2017) Curcumin inhibits tau aggregation and disintegrates
preformed tau filaments in vitro. J. Alzheimer’s Disease. 60: 999-1014
Sawant, AV, Srivastave, S, Prassanawar, SS, Bhattacharyya, B and Panda, D (2019) Crocin, a
carotenoid, suppresses spindle microtubule dynamics and activates the mitotic checkpoint
by binding to tubulin. Biochem. Pharmacol. 163: 32-45.
30