Identification of 5-benzylidene-2-phenylthiazolones as potent PRMT5 inhibitors by virtual screening, structural optimization and biological evaluations

Article abstract of DOI:10.1016/j.bioorg.2018.08.021

Protein arginine methyltransferase 5 (PRMT5) is an epigenetics related enzyme that has been validated as an important therapeutic target for glioblastoma and mantel cell lymphoma. In the present study, 11 novel PRMT5 inhibitors with 5-benzylidene-2-phenyl

Full text of DOI:10.1016/j.bioorg.2018.08.021

Accepted Manuscript
Identification of 5-Benzylidene-2-phenylthiazolones as Potent PRMT5 Inhibi-
tors by Virtual Screening, Structural Optimization and Biological Evaluations
Kongkai Zhu, Hongrui Tao, Jia-Li Song, Lu Jin, Yuanyuan Zhang, Jingqiu Liu,
Zhifeng Chen, Cheng-Shi Jiang, Cheng Luo, Hua Zhang
PII:
S0045-2068(18)30654-0
DOI:
https://doi.org/10.1016/j.bioorg.2018.08.021
YBIOO 2478
Reference:
Bioorganic Chemistry
To appear in:
Received Date:
Revised Date:
Accepted Date:
2 July 2018
13 August 2018
14 August 2018
Please cite this article as: K. Zhu, H. Tao, J-L. Song, L. Jin, Y. Zhang, J. Liu, Z. Chen, C-S. Jiang, C. Luo, H. Zhang,
Identification of 5-Benzylidene-2-phenylthiazolones as Potent PRMT5 Inhibitors by Virtual Screening, Structural
Optimization and Biological Evaluations, Bioorganic Chemistry (2018), doi: https://doi.org/10.1016/j.bioorg.
2018.08.021
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Identification of 5-Benzylidene-2-phenylthiazolones as Potent
PRMT5 Inhibitors by Virtual Screening, Structural
Optimization and Biological Evaluations
Kongkai Zhu$,|, Hongrui Tao$,†,|, Jia-Li Song$,|, Lu Jin#, Yuanyuan Zhang†, Jingqiu
Liu†, Zhifeng Chen†, Cheng-Shi Jiang$,*, Cheng Luo†,* and Hua Zhang$,*
$School of Biological Science and Technology, University of Jinan, Jinan 250022,
P.R. China;
#Institute of Pharmacology of Natural Products & Clinical Pharmacology, Ulm
University, Germany;
† Drug Discovery and Design Center, State Key Laboratory of Drug Research,
Shanghai Institute of Materia Medica, Chinese Academy of Sciences, Shanghai
201203, P.R. China
*Correspondence: Cheng-Shi Jiang, E-mail: jiangchengshi-20@163.com; Cheng Luo, E-mail:
cluo@simm.ac.cn; Hua Zhang, E-mail: bio_zhangh@ujn.edu.cn
| These authors contributed equally.
1

Abstract
Protein arginine methyltransferase 5 (PRMT5) is an epigenetics related enzyme that
has been validated as an important therapeutic target for glioblastoma and mantel cell
lymphoma. In the present study, 11 novel PRMT5 inhibitors with
5-benzylidene-2-phenylthiazolone scaffold were identified by molecular
docking-based virtual screeningand structural optimization. Their IC50 values against
PRMT5 at enzymatic level were ranging from 0.77 to 23 μM. As expected, the top
two active hits (5 and 19) showed potent anti-proliferative activity against MV4-11
cells with EC50 values lower than 10 μM and reduced the cellular symmetric arginine
dimethylation levels of SmD3 protein. Besides, 5 and 19 demonstrated the mechanism
of cell killing in cell cycle arrest and apoptotic effect. The probable binding modes of
the two compounds were explored and further verified by molecular dynamics
simulation. The structure-activity relationship (SAR) of this class of structures was
also discussed and further demonstrated by molecular docking simulation.
Keywords: PRMT5 inhibitor; arginine methylation; virtual screening; molecular
docking, molecular dynamics simulation
Introduction
Protein arginine methyltransferase 5 (PRMT5) belongs to an important family of
enzymes that mediate gene transcription and cellular signaling by catalyzing the
methylation of arginine residues of their substrates.1 Using S-adenosyl-L-methionine
(SAM) as the methyl donor, PRMT5 catalyzes the symmetric dimethylation of its
histone and non-histone substrates assisted by its partner WD repeat protein MEP50
(methylosome protein 50, also known as Wdr77).2 PRMT5 can silence the
transcription of regulatory genes by methylating the H3 residue Arg8 (H3R8me2S)
and H4 residue Arg3 (H4R3me2S) of histones.3, 4 Recently, studies have indicated that
PRMT5 is a promising anticancer target for many human malignancies.5-8 For
example, upregulation of PRMT5 was observed in mantle cell lymphoma (MCL)
patient samples, and knockdown of PRMT5 elicited potent antiproliferative effects in
2

MCL cell lines.4, 9-11 In addition, genetic attenuation of PRMT5 resulted in cell-cycle
arrest, apoptosis, and loss of cell migratory activity of glioblastoma cells.7 In lung
cancer cells, PRMT5 was reported to promote cell proliferation by regulating multiple
signaling pathways.6 As a consequence, lots of efforts have been made to develop
modulators of PRMT5. Encouragingly, one PRMT5 inhibitor GSK-3326595 (Fig. 1)
has been put into phase I clinical trial to study the dose escalation in subjects with
solid tumors (glioblastoma, triple negative breast cancer and metastatic transitional
cell carcinoma of the bladder) and non-Hodgkin's lymphoma. However, limited
numbers of PRMT5 inhibitors with only five different scaffolds (Fig. 1) have been
discovered so far. Considering the relatively poor druggability and lack of selectivity
of the reported PRMT5 inhibitors, there is an urgent need to discover more novel
PRMT5 inhibitors.
As shown in the crystal structure of human PRMT5 (Fig. S1), there are two binding
sites (SAM and substrate binding sites) that can be used for the discovery of new
inhibitors. Therefore, two strategies can be applied to perform structure-based virtual
screening; one is targeting SAM binding site and the other is targeting substrate
binding site. Accordingly, the previously reported PRMT5 inhibitors can be divided
into two types. For example, DC_Y13412 and the pan-methyltransferase inhibitor
sinefungin13(Fig. 1) belong to the SAM binding site group, while GSK-3326595,
EPZ0156665 and DC-C0114 (Fig. 1) pertain to the substrate binding site class.
Although it is difficult to identify potent inhibitors that could compete with SAM,
many attempts have been made to identify inhibitors that could bind to SAM binding
site, and two scaffolds (1715 and P5i-616) of non-SAM analogues have been reported
up to now.
In the present work, a novel small-molecule PRMT5 inhibitor (P16) with
5-benzylidene-2-phenylthiazolone scaffold was initially identified by molecular
docking-based virtual screening and showed an IC50 value of 13.9 μM in the bioassay.
Further substructure search in SPECS database and structural optimization from P16
3

returned 44 extra analogues, ten of which displayed promising PRMT5 inhibitory
activity with IC50 values ranging from 0.77 to 23 μM. The structure-activity
relationship (SAR) of these structures was elucidated. Besides, the probable binding
modes of the top two active compounds 5 and 19 were explored by molecular docking
and molecular dynamics simulations. Moreover, 5 and 19 also exhibited significant
anti-proliferative activity against MV4-11 leukemia cells and altered the endogenous
arginine methylation level of SmD3, a protein that could be used to track the cell
biochemical activity of PRMT5.17-19 The scaffold discovered in this work can be
further optimized and used as a chemical probe to investigate the functions of
PRMT5.
Results and discussion
Molecular docking-based virtual screening leading to the discovery of the hit
compound P16.
Currently, ten X-ray crystal structures of human PRMT5 are available in RCSB
protein data bank (PDB) with codes 4X60,5 4X61,5 4X63,5 4GQB13 5C9Z, 5EMJ,20
5EMK,20 5EML,20 5EMM,20 and 5FA5,21 which enables the feasibility of molecular
docking-based virtual screening. In order to build an optimum structural model to
perform molecular docking, these crystal structures were carefully analyzed and a
model containing PRMT5:MEP50, SAM and H4 peptide was finally constructed.
Briefly, the crystal structure with PDB code 5EML comprising the cofactor SAM was
chosen as template to build the PRMT5:MEP50 and SAM fragments, while 4GQB
which was the only one containing a H4 peptide in substrate binding site was chosen
as the template to construct the H4 peptide part. In the constructed structure model,
the coordinates of PRMT5:MEP50 and SAM were originated from 5EML while the
H4 peptide was originated from 4GQB. Docking SAM to its binding site could well
reproduce the binding mode resolved by the crystal structure, which supported the
reliability of the current virtual screening method. In addition, a small library
containing 20 small-molecule PRMT5 inhibitors (analogues of 17 and P5i-6, IC50 <
10 μM) and 780 decoys were manually prepared to test the enrichment capability of
4

the docking method. All the 20 active compounds were found in top 10% range
ranked by XP Gscore when screening this library, which further validated the
reliability and effectiveness of the method.
SPECS database (http://www.specs.net) containing 212,255 compounds were
screened in silico to obtain candidates that showed potent binding affinity with
PRMT5, and the top 1000 compounds ranked by XP Gscore were retained for visual
inspection to retrieve compounds possessing well chemical fitness with SAM binding
site and diverse chemical scaffolds. After filtration by “Lipinski’s Rule of Five” and
clustering with “Clustering Molecules” protocol by Pipeline Pilot 7.5, 40 molecules
were selected and purchased for subsequent biological evaluation at enzymatic level
(Fig. 2A, the virtual screening workflow). Among them, one hit compound (P16)
displayed the best inhibitory activity (Table S1) when tested by the Alpha LISA
method15 with an IC50 value of 13.9 μM (Fig. 2B).
Hit optimization returned the submicromolar active lead compound 5
Similarity-based analogue search and synthetic structural optimization were carried
out to improve the potency of the hit compound P16. Substructure search method was
used to do similarity-based analogue search and retrieved 19 analogues in SPECS
database. Meanwhile, structural optimization afforded 25 additional analogues. The
latter 25 compounds were smoothly synthesized by one-pot reaction according to a
previously reported protocol22 using the reagents including 4-methylbenzonitrile,
2-mercaptoacetic acid and substituted benzaldehydes in hand. The synthetic route was
outlined in Scheme 1. The double bond configuration of the synthesized compounds
was assigned to be Z based NMR and HPLC analyses and similar structural analogues
were also reported with Z geometry in the literature.22
Scheme 1. Synthesis of thiazolone-based P16 analogues. Reagents and conditions:
5

Et3N, MeOH or EtOH, reflux, 12 h.
The total of 44 compounds was then tested in vitro for their inhibitory activities
against PRMT5. The inhibition rates at 50 μM were first obtained and the IC50 values
of those with >40% inhibition rate were further determined (Table 1). Alpha LISA
assay identified ten active compounds with IC50 values ranging from 0.77 to 23 μM.
The structure-activity relationship (SAR) of this series of compounds was discussed
below.
Clearly, compounds showing inhibition rate above 40% at 50 μM were found to have
at least one hydroxyl group on the benzene ring (R2), and those without the hydroxyl
substituent were inactive. This obviously indicated that the presence of hydroxyl
group as hydrogen bond donor or acceptor played a key role in their activity, which
was lately validated by the predicted binding mode from molecular docking. In
addition, comparison of activities between the compounds with R1 as methyl and as H
suggested that the presence of methyl (R1) was not necessary. For example, each pair
of active compounds 14/41, 22/44 and 42/P16 showed similar activity toward PRMT5.
In most cases, the alkylation of hydroxyl(s) at meta- and/or ortho-position of benzene
ring reduced the activity, which was reflected by the inhibition rates of compounds
3-18. It was worth noting that the introduction of a halogen element (e.g., Cl or Br) in
14, 16, 20 and 41 increased their PRMT5 inhibitory activity compared with 6,
suggesting that the halogen atom was also vital to the activity in the condition of the
presence of hydroxyl group(s). Meanwhile, substitution of Cl and Br seemed to have a
similar effect on the compounds’ activity (16 vs 20).
Next, the IC50 values of compounds 5, 14-16, 19, 20, 22, 41, 42, and 44 for PRMT5
inhibition were tested. Most of the compounds showed the same level activity to P16
with IC50 values ranging from 11.0 to 23.0 μM. Among them, two analogues 5 and 19
displayed significantly improved activity with IC50 values of 0.77 and 6.60 μM,
respectively (Fig. 3A).
6

Compound 5 showed selective inhibitory activity against PRMT5.
The most active 5 was chosen for further test on its inhibition against the other key
homological member of PRMT5, PRMT1. As shown in Fig. S2, the assay results
(IC50 = 10.0 μM) revealed that 5 was displayed moderate selectivity (13-fold) against
PRMT5.
Predicted binding modes with PRMT5 of the top two active compounds
Compounds 5 and 19 showed the best inhibitory activity against PRMT5 among all
tested analogues, and their binding modes with PRMT5 were thus analyzed. As shown
in Fig. 3B–3C, the two compounds showed similar binding modes with quite
consistent space orientations. Hydrophobic, hydrogen bond, cation-π, and π-π
interactions were observed and contributed to the binding affinity by further analysis
(Fig. 4). Both compounds 5 and 19 were located in the hydrophobic pocket which is
composed of residues Y324, L319, P314, L315, L436, M420 and A366 (Fig. 4). In
particular, the methylbenzene ring established cation−π interaction with the side chain
of K393. Not surprisingly, all hydroxyl groups in the two compounds formed
hydrogen bonds with residues G365 of PRMT5, indicative of their important
contribution to the activity. By comparing the binding modes of the two compounds
with that of SAM (Fig. S3), we concluded that this series of compounds could
efficiently occupy the SAM binding pocket of PRMT5.
In order to further validate their binding modes, 100 ns molecular dynamics
simulations were conducted on the two binding poses. The interactions between each
compound and PRMT5 were found to be stable during the MD simulation process by
analyzing the root mean square deviation (RMSD) values of each MD trajectory (Fig.
S4 and S5). As the two compounds showed equivalent XP Gscore, the binding free
energies of 5 and 19 with PRMT5 were calculated by MM/PBSA method with values
of –31.85 kcal/mol and –27.34 kcal/mol, respectively. The results were in well
accordance with their experimental activity data.
7

Anti-proliferative and cellular symmetric dimethylation effects of 5 and 19 on
leukemia cell line MV4-11
Leukemia cell line MV4-11 was used to measure the anti-proliferative effects of 5 and
19, and the results were summarized in Fig. 5A. The two compounds showed a dose-
and time-dependent manner in inhibiting MV4-11 cell proliferation with 12-day EC50
values of 7.93 μM and 3.30 μM, respectively. It was interesting to note that 5
displayed more potent inhibitory activity than 19 at enzymatic level but was less
active at cell level. This observation could result from the higher cell membrane
permeability of 19 as indicated by its better ClogP parameter (2.50) than that (4.03) of
5.
The effects of 5, 19 and EPZ015666 (positive control) on cellular symmetric arginine
dimethylation in MV4-11 cells (Fig. 5B and Fig. S6) were tested by immunoblot
using symmetric dimethyl arginine (SDMA) antibody. Similarly to EPZ015666,
treatment of 5 and 19 led to a concentration dependent decrease in the intensity of
multiple bands, including PRMT5 substrate SmD3 protein. Concentration-dependent
decreases in SmD3me2s were observed, which demonstrated the PRMT5 engagement
effects of the compounds in a cellular context.
Inducement of cell cycle arrest and cell apoptosis of 5 and 19
The effects of 5 and 19 on cell cycle and cell apoptosis were examined to illustrate
their anti-proliferation mechanism through flow cytometric analysis. The results
showed a concentration-dependent cell cycle arrest at G1 phase (Fig. 6 and Fig. S7)
and increased apoptosis (Fig. 7 and Fig. S8) upon treatment with the two compounds
for 24 and 48 hours, respectively. These results were in agreement with the critical
role of PRMT5 in glioblastoma differentiated cells (GBMDC) and glioblastoma
neurospheres (GBMNS), which uncovered that PRMT5 knockdown in GBMDC led
to cell apoptosis and knockdown in GBMNS led to G1 cell cycle arrest.23 These
findings indicated that this series of compounds were both cell-cycle arrestors and
8

apoptosis inducers.
Experimental methods
Molecular docking-based virtual screening
Among the ten determined X-ray crystal structures of human PRMT5, 5EML and
4X61 are the two structures containing the cofactor SAM, while 4GQB is the only
one that comprises a histone H4 peptide in substrate binding site with a PRMT5
inhibitor occupying SAM binding site. Therefore, 4GQB and 5EML with higher
resolution (2.39 Å) than 4X61 (2.85 Å) were chosen to construct the structure model
for virtual screening targeting SAM binding site. By superposing 5EML with 4GQB
and replacing the coordinates of the substrate binding site inhibitor 5QK in 5EML
with H4 peptide in 4GQB, a new complex model containing PRMT5:MEP50, SAM
and H4 peptide was finally obtained and used for further study. In this constructed
complex model, all the coordinates were derived from 5EML except for the H4
peptide. The docking simulation was performed using Schrödinger software running
under Maestro version 7.5. Glide program24, 25 embedded in Maestro 7.5 was used to
carry out molecular docking. During the docking process, the coordinates of the
protein (PRMT5:MEP50 and H4 peptide) were first minimized using the Protein
Preparation Wizard Workflow with default settings, and docking grids were then
created by defining residues within 15 Å around SAM. Finally, compounds of SPECS
database prepared by the LigPrep panel (version , chr dinger, LLC, New York,
NY) were docked into the well-defined docking grids with the extra precision (XP)
mode.
Chemistry
All purchased compounds were from commercial supplier (SPECS database) and used
without further purification. Purity of these compounds was >= 95% as declared by
the chemical vendor (Table S1–S2).
General methods. Melting points were measured on a RY-1 melting point apparatus
9

(Tianjin Analysis Instrument Factory, Tianjin, China). Commercially available
reagents were used without further purification. Organic solvents were evaporated
under reduced pressure using a Büchi R-100 evaporator. Reactions were monitored by
TLC using GF254 pre-coated silica gel plates (Yantai JiangYou Silica Gel Inc., Yantai,
China). Column chromatography was performed on 200–300 mesh silica gel
(Qingdao Marine Chemical Inc., Qingdao, China). NMR spectra were acquired on a
Bruker AVANCE DRX600 spectrometer. Chemical shifts were expressed in  (ppm)
and coupling constants (J) in Hz using solvent signals as references (CDCl3, H 7.26
ppm and C 77.23 ppm; CD3OD, H 3.31 ppm and C 49.00 ppm; DMSO-d6, H 2.50
ppm and C 39.50 ppm). The purity of the samples was determined by analytical
HPLC using an Agilent 1260 system with ZDRBAX SB-C18 column (4.6 mm×150
mm). HPLC data were recorded using parameters as follows: H2O/MeOH, 60/40 to
0/100 in 30 min, +30 min isocratic, flow rate of 1.0 mL/min,  = 280 and 320 nm.
High-resolution mass spectra (HRMS) were recorded on an Agilent Q-TOF 6520 MS
spectrometer.
General procedure for synthesis of 5-benzylidene-2-phenylthiazolones 1-8, 10, 11, and
23-37. A solution of p-methylbenzonitrile (117 mg, 1 mmol), thioglycolic acid (101
mg, 1.1 mmol), corresponding benzaldehyde (1 mmol) and 0.5 mL triethylamine in 5
mL methanol or ethanol was refluxed for at least 12 h. The reaction mixture was
evaporated under reduced pressure, and the residue was purified by silica gel column
chromatography to afford the desired product.
(Z)-5-Benzylidene-2-(p-tolyl)thiazol-4(5H)-one (1)
1
A yellow solid (46 mg, 33%). mp 190–192 °C; HPLC purity: 17.74 min, 95.2%. H
NMR (600 MHz, CDCl3) δ 8.12 (d, J = 7.8 Hz, 2H), 8.05 (s, 1H), 7.68 (d, J = 8.4 Hz,
2H), 7.51 (dd, J = 7.1, 8.0 Hz, 2H), 7.47 (d, J = 7.1 Hz, 1H), 7.37 (d, J = 8.0 Hz, 2H),
2.48 (s, 3H). 13C NMR (150 MHz, CDCl3) δ 187.3, 183.5, 146.9, 138.2, 134.0, 131.1,
130.8, 130.1, 129.4, 129.3, 129.1, 126.7, 22.2. HRMS (ESI+): calcd for [M+H]+
C17H14NOS+ m/z 280.0791; found 280.0795.
10

(Z)-5-(2-Methoxybenzylidene)-2-(p-tolyl)thiazol-4(5H)-one (2)
1
A yellow solid (57 mg, 37%). mp 166–168 °C; HPLC purity: 17.77 min, 98.8%. H
NMR (600 MHz, DMSO-d6) δ 8.23 (s, 1H), 8.11 (d, J = 8.2 Hz, 2H), 7.69 (dd, J = 1.2,
7.7 Hz, 1H), 7.55 (ddd, J = 1.5, 8.6, 8.6 Hz, 1H), 7.48 (d, J = 8.2 Hz, 2H), 7.20 (d, J =
8.4 Hz, 1H), 7.16 (dd, J = 7.5, 7.6 Hz, 1H), 3.94 (s, 3H), 2.45 (s, 3H). 13C NMR (150
MHz, DMSO-d6) δ 187.4, 182.2, 158.7, 146.8, 133.4, 131.7, 130.2, 129.1, 128.6,
128.6, 126.1, 121.9, 121.1, 112.1, 56.0, 21.5. HRMS (ESI+): calcd for [M+H]+
C18H16NO2S+ m/z 310.0896; found 31.0893.
(Z)-5-(3-Methoxybenzylidene)-2-(p-tolyl)thiazol-4(5H)-one (3)
1
A yellow solid (66 mg, 43%). mp 158–160 °C; HPLC purity: 17.94 min, 98.1%. H
NMR (600 MHz, CDCl3) δ 8.12 (d, J = 8.2 Hz, 2H), 8.01 (s, 1H), 7.42 (dd, J = 7.8,
8.2 Hz, 1H), 7.37 (d, J = 8.2 Hz, 2H), 7.28 (d, J = 7.8 Hz, 1H), 7.19 (d, J = 2.4 Hz,
13
1H), 7.02 (dd, J = 2.4, 8.2 Hz, 1H), 3.89 (s, 3H), 2.48 (s, 3H). C NMR (150 MHz,
CDCl3) δ 187.3, 183.5 160.2 146.9, 138.1 135.4 130.4 130.1, 129.3, 129.1, 127.0,
123.4, 116.9, 115.8, 55.6, 22.2. ESIMS m/z: 310.0 [M + H]+. HRMS (ESI+): calcd for
[M+H]+ C18H16NO2S+ m/z 310.0896; found 310.0890.
(Z)-5-(4-Methoxybenzylidene)-2-(p-tolyl)thiazol-4(5H)-one (4)
1
An orange solid (53 mg, 34%). mp 221–223 °C; HPLC purity: 17.74 min, 95.4%. H
NMR (600 MHz, CDCl3) δ 8.11 (d, J = 8.2 Hz, 2H), 8.02 (s, 1H), 7.65 (d, J = 8.6 Hz,
2H), 7.36 (d, J = 8.0 Hz, 2H), 7.02 (d, J = 8.6 Hz, 2H), 3.89 (s, 3H), 2.48 (s, 3H). 13C
NMR (150 MHz, CDCl3) δ 186.6, 183.8, 162.1, 146.5, 138.3, 132.0, 130.1, 129.5,
129.0, 126.7, 123.9, 115.0, 55.7, 22.1. HRMS (ESI+): calcd for [M+H]+ C18H16NO2S+
m/z 310.0896; found 310.0893.
(Z)-5-(3,4-Dihydroxybenzylidene)-2-(p-tolyl)thiazol-4(5H)-one (5)
1
A dark red solid (34 mg, 22%). mp 256–258 °C; HPLC purity: 15.19 min, 96.4%. H
NMR (600 MHz, MeOD) δ 8.11 (d, J = 8.2 Hz, 2H), 7.91 (s, 1H), 7.45 (d, J = 8.2 Hz,
2H), 7.24 (d, J = 2.4 Hz, 1H), 7.18 (dd, J = 8.2, 2.4 Hz, 1H), 6.92 (d, J = 8.2 Hz, 1H),
2.48 (s, 3H). 13C NMR (150 MHz, MeOD) δ 188.6, 185.1, 1513, 148.2, 147.4, 141.2,
131.2, 130.4, 129.7, 126.9, 126.8, 123.1, 118.0, 117.1, 21.9. HRMS (ESI+): calcd for
[M+H]+ C17H14NO3S+ m/z 312.0689; found 312.0686.
11

(Z)-5-(4-Hydroxy-3-methoxybenzylidene)-2-(p-tolyl)thiazol-4(5H)-one (6)
A red-orange solid (42 mg, 26%). mp 197–199 °C; HPLC purity: 16.26 min, 95.9%.
1H NMR (600 MHz, CDCl3) δ 8.10 (d, J = 8.0 Hz, 2H), 7.98 (s, 1H), 7.36 (d, J = 8.0
Hz, 2H), 7.29 (d, J = 8.2 Hz, 1H), 7.15 (d, J = 1.8 Hz, 1H), 7.04 (d, J = 8.2 Hz, 1H),
6.08 (brs, OH), 4.00 (s, 3H), 2.48 (s, 3H). 13C NMR (150 MHz, CDCl3) δ 186.4, 183.7,
148.9, 147.1, 146.6, 138.7, 130.1, 129.5, 129.0, 126.6, 125.8, 123.8, 115.4, 112.8,
56.2, 22.1. ESIMS m/z: 326.0 [M + H]+. HRMS (ESI+): calcd for [M+H]+
C18H16NO3S+ m/z 326.0845; found 326.0846.
(Z)-5-(3,5-dimethoxybenzylidene)-2-(p-tolyl)thiazol-4(5H)-one (7)
1
A yellow solid (64 mg, 38%). mp 184–186 °C; HPLC purity: 18.18 min, 91.7%. H
NMR (600 MHz, CDCl3) δ 8.10 (d, J = 8.2 Hz, 2H), 7.94 (s, 1H), 7.36 (d, J = 8.2 Hz,
2H), 6.80 (d, J = 2.1 Hz, 1H), 6.80 (d, J = 2.1 Hz, 1H), 6.56 (dd, J = 2.1 Hz, 1H), 3.86
(s, 6H), 2.47 (s, 3H). 13C NMR (150 MHz, CDCl3) δ 187.3, 183.4, 161.3, 146.9, 138.2,
135.8, 130.1, 129.3, 129.1, 127.2, 108.6, 103.2, 55.7, 22.1. ESIMS m/z: 340.0 [M +
H]+. HRMS (ESI+): calcd for [M+H]+ C19H18NO3S+ m/z 340.1002; found 340.1008.
(Z)-5-(3,4-Dimethoxybenzylidene)-2-(p-tolyl)thiazol-4(5H)-one (8)
A yellow-orange solid (44 mg, 26%). mp 187–189 °C; HPLC purity: 17.08 min,
95.7%. 1H NMR (600 MHz, CDCl3) δ 8.11 (d, J = 8.1 Hz, 2H), 8.00 (s, 1H), 7.36 (d, J
= 8.1 Hz, 2H), 7.32 (dd, J = 1.7, 8.4 Hz, 1H), 7.18 (d, J = 1.7 Hz, 1H), 6.99 (d, J = 8.4
Hz, 1H), 3.99 (s, 3H), 3.97 (s, 3H), 2.48 (s, 3H). 13C NMR (150 MHz, CDCl3) δ 186.4,
183.7, 150.9, 149.6, 146.6, 138.6, 130.1, 129.5, 129.0, 127.0, 125.6, 124.1, 112.9,
111.6, 56.3, 56.2, 22.2. ESIMS m/z: 340.0 [M + H]+. HRMS (ESI+): calcd for [M+H]+
C19H18NO3S+ m/z 340.1002; found 340.1004.
(Z)-5-(3-Hydroxy-4,5-dimethoxybenzylidene)-2-(p-tolyl)thiazol-4(5H)-one (10)
1
An orange solid (59 mg, 33%). mp 179–181 °C; HPLC purity: 16.33 min, 96.0%. H
NMR (600 MHz, CDCl3) δ 8.10 (d, J = 8.2 Hz, 2H), 7.91 (s, 1H), 7.36 (d, J = 8.2 Hz,
2H), 7.00 (d, J = 1.9 Hz, 1H), 6.77 (d, J = 1.9 Hz, 1H), 5.97 (brs, OH), 3.99 (s, 3H),
3.94 (s, 3H), 2.48 (s, 3H). 13C NMR (150 MHz, CDCl3) δ 187.0, 183.6, 152.6, 149.8,
146.8, 138.2, 138.1, 130.1, 129.8, 129.4, 129.1, 126.0, 110.1, 107.8, 61.3, 56.2, 22.1.
HRMS (ESI+): calcd for [M+H]+ C19H18NO4S+ m/z 356.0951; found 356.0959.
12

(Z)-5-(4-Hydroxy-3,5-dimethoxybenzylidene)-2-(p-tolyl)thiazol-4(5H)-one (11)
A red-orange solid (50 mg, 28%). mp 227–229 °C; HPLC purity: 16.14 min, 98.5%.
1H NMR (600 MHz, CDCl3) δ 8.10 (d, J = 8.1 Hz, 2H), 7.96 (s, 1H), 7.36 (d, J = 8.1
Hz, 2H), 6.93 (s, 2H), 5.98 (brs, OH), 3.99 (s, 6H), 2.48 (s, 3H). 13C NMR (150 MHz,
CDCl3) δ 186.3, 183.6, 147.6, 146.6, 138.9, 138.2, 130.1, 129.4, 129.0, 125.6, 124.1,
108.1, 56.6, 22.2. HRMS (ESI+): calcd for [M+H]+ C19H18NO4S+ m/z 356.0951; found
356.0946.
(Z)-5-(4-Isopropylbenzylidene)-2-(p-tolyl)thiazol-4(5H)-one (23)
1
A yellow solid (59 mg, 37%). mp 146–148 °C; HPLC purity: 19.29 min, 91.0%. H
NMR (600 MHz, CDCl3) δ 8.11 (d, J = 8.2 Hz, 2H), 8.04 (s, 1H), 7.62 (d, J = 8.2 Hz,
13
2H), 7.37 (m, 4H), 2.98 (m, 1H), 2.48 (s, 3H), 1.29 (d, J = 6.9 Hz, 6H). C NMR
(150 MHz, CDCl3) δ 187.0, 183.7, 152.8, 146.7, 138.4, 131.6, 131.1, 130.1, 129.4,
129.0, 127.6, 125.6, 34.4, 23.8, 22.1. HRMS (ESI+): calcd for [M+H]+ C20H20NOS+
m/z 322.1260; found 322.1268.
(Z)-5-(4-(Dimethylamino)benzylidene)-2-(p-tolyl)thiazol-4(5H)-one (24)
1
A purple solid (63 mg, 36%). mp 161–163 °C; HPLC purity: 18.00 min, 97.8%. H
NMR (600 MHz, CDCl3) δ 8.10 (d, J = 8.4 Hz, 2H), 8.00 (s, 1H), 7.59 (d, J = 8.9 Hz,
2H), 7.33 (d, J = 8.2 Hz, 2H), 6.74 (d, J = 8.9 Hz, 2H), 3.10 (s, 6H), 2.46 (s, 3H), 1.22
(t, J = 7.2 Hz, 6H). 13C NMR (150 MHz, CDCl3) δ 185.2, 184.1, 152.2, 145.8, 139.7,
133.4, 123.0, 129.8, 128.7, 121.6, 120.1, 112.1, 40.2, 22.0. HRMS (ESI+): calcd for
[M+H]+ C19H19N2OS+ m/z 323.1213; found 323.1208.
(Z)-5-(4-(Diethylamino)benzylidene)-2-(p-tolyl)thiazol-4(5H)-one (25)
1
A purple solid (43 mg, 27%). mp 222–224 °C; HPLC purity: 18.79 min, 99.5%. H
NMR (600 MHz, CDCl3) δ 8.09 (d, J = 8.0 Hz, 2H), 7.97 (s, 1H), 7.56 (d, J = 8.8 Hz,
2H), 7.33 (d, J = 8.0 Hz, 2H), 6.72 (d, J = 8.8 Hz, 2H), 3.44 (q, J = 7.1 Hz, 4H), 2.45
13
(s, 6H), 1.22 (t, J = 7.1 Hz, 6H). C NMR (150 MHz, CDCl3) δ 184.9, 184.1, 150.1,
145.7, 139.7, 133.7, 129.9, 129.9, 128.6, 120.8, 119.4, 111.7, 44.8, 22.0, 12.7. HRMS
(ESI+): calcd for [M+H]+ C21H23N2OS+ m/z 351.1526; found 351.1531.
(Z)-5-(3,4-Dimethylbenzylidene)-2-(p-tolyl)thiazol-4(5H)-one (26)
1
A yellow solid (38 mg, 25%). mp 200–202 °C; HPLC purity: 19.08 min, 92.1%. H
13

NMR (600 MHz, CDCl3) δ 8.12 (d, J = 8.2 Hz, 2H), 8.01 (s, 1H), 7.44 (s, 1H), 7.43 (d,
J = 6.4 Hz, 1H), 7.36 (d, J = 8.2 Hz, 2H), 7.26 (d, J = 6.4 Hz, 1H), 2.48 (s, 3H), 2.35
(s, 3H), 2.34 (s, 3H). 13C NMR (150 MHz, CDCl3) δ 187.0, 183f.7, 146.6, 140.9,
138.8, 137.8, 132.1, 131.7, 130.0, 130.1, 129.5, 129.0, 128.5, 125.3, 22.2, 20.2, 20.0.
HRMS (ESI+): calcd for [M+H]+ C19H18NOS+ m/z 308.1104; found 308.1100.
(Z)-5-(4-Methylbenzylidene)-2-(p-tolyl)thiazol-4(5H)-one (27)
1
A yellow solid (35 mg, 24%). mp 230–232 °C; HPLC purity: 18.47 min, 92.3%. H
NMR (600 MHz, CDCl3) δ 8.11 (d, J = 8.0 Hz, 2H), 8.03 (s, 1H), 7.58 (d, J = 7.8 Hz,
2H), 7.36 (d, J = 8.0 Hz, 2H), 7.31 (d, J = 7.8 Hz, 2H), 2.48 (s, 3H), 2.43 (s, 3H). 13C
NMR (150 MHz, CDCl3) δ 187.0, 183.7, 146.7, 142.0, 138.4, 131.3, 130.9, 130.2,
130.1, 129.4, 129.0, 125.5, 22.1, 21.8. HRMS (ESI+): calcd for [M+H]+ C18H16N2OS+
m/z 294.0947; found 294.0950.
(Z)-5-(3-Fluoro-4-methylbenzylidene)-2-(p-tolyl)thiazol-4(5H)-one (28)
A yellow-orange solid (70 mg, 45%); HPLC purity: 18.60 min, 91.4%. mp
230–232 °C. 1H NMR (600 MHz, CDCl3) δ 8.12 (d, J = 8.4 Hz, 2H), 7.97 (s, 1H),
13
7.38 (d, J = 8.4 Hz, 2H), 7.33 (m, 3H), 2.49 (s, 3H), 2.35 (d, J = 1.2 Hz, 3H). C
NMR (150 MHz, CDCl3) δ 187.0, 183.5, 161.7 (d, J = 247 Hz), 147.0, 136.9, 136.9,
133.6 (d, J = 7.8 Hz), 132.4 (d, J = 5.5 Hz), 130.2, 129.3, 129.2, 126.9 (d, J = 25.4
Hz), 126.8, 116.4 (d, J = 23.5 Hz), 22.8, 15.0. HRMS (ESI+): calcd for [M+H]+
C18H15FNOS+ m/z 312.0853; found 312.0845.
(Z)-5-(4-Fluoro-3-methylbenzylidene)-2-(p-tolyl)thiazol-4(5H)-one (29)
1
A yellow solid (59 mg, 38%). mp 209–211 °C; HPLC purity: 18.47 min, 92.8%. H
NMR (600 MHz, CDCl3) δ 8.11 (d, J = 8.2 Hz, 2H), 7.97 (s, 1H), 7.51-7.50 (m, 2H),
7.37 (d, J = 8.2 Hz, 2H), 7.13 (dd, J = 9.1, 9.1 Hz, 1H), 2.48 (s, 3H), 2.37 (s, 3H). 13C
NMR (150 MHz, CDCl3) δ 187.0, 183.5, 162.9 (d, J = 254 Hz), 146.9, 137.3, 134.2 (d,
J = 6.0 Hz), 130.3 (d, J = 5.7 Hz), 130.1, 130.1 (d, J = 3.7 Hz), 129.3, 129.1, 126.4 (d,
J = 17.4 Hz), 126.0 (d, J = 2.5 Hz), 116.4 (d, J = 13.2 Hz), 22.2, 14.8 (d, J = 3.3 Hz).
HRMS (ESI+): calcd for [M+H]+ C18H15FNOS+ m/z 312.0853; found 312.0855.
(Z)-2-(p-Tolyl)-5-(3-(trifluoromethyl)benzylidene)thiazol-4(5H)-one (30)
1
An orange solid (59 mg, 34%). mp 196–198 °C; HPLC purity: 17.75 min, 93.2%. H
14

NMR (600 MHz, CDCl3) δ 8.13 (d, J = 8.1 Hz, 2H), 8.03 (s, 1H), 7.90 (s, 1H), 7.84 (d,
J = 7.7 Hz, 1H), 7.71 (d, J = 7.7 Hz, 1H), 7.64 (dd, J = 7.7, 7.7 Hz, 1H), 7.38 (d, J =
8.1 Hz, 2H), 2.49 (s, 3H). 13C NMR (150 MHz, CDCl3) δ 187.2, 183.1, 147.3, 135.81,
134.9, 133.6, 132.0 (q, J = 32.8 Hz), 130.2, 130.0, 129.3, 129.1, 128.7, 127.3 (q, J =
3.8 Hz), 127.0 (q, J = 3.6 Hz), 123.8 (q, J = 273 Hz), 22.2. HRMS (ESI+): calcd for
[M+H]+ C18H13F3NOS+ m/z 348.0664; found 348.0674.
(Z)-5-(4-Chlorobenzylidene)-2-(p-tolyl)thiazol-4(5H)-one (31)
1
A brown solid (47 mg, 30%). mp 187–189 °C; HPLC purity: 17.78 min, 92.0%. H
NMR (600 MHz, CDCl3) δ 8.11 (d, J = 8.2 Hz, 2H), 7.98 (s, 1H), 7.61 (d, J = 8.5 Hz,
13
2H), 7.48 (d, J = 8.5 Hz, 2H), 7.37 (d, J = 8.2 Hz, 2H), 2.48 (s, 3H). C NMR (150
MHz, CDCl3) δ 186.8, 183.2, 146.9, 137.0, 136.4, 132.4, 131.7, 130.1, 129.6, 129.1,
129.0, 127.1, 22.0. HRMS (ESI+): calcd for [M+H]+ C17H13ClNOS+ m/z 314.0401;
found 314.0398.
(Z)-5-(4-Bromobenzylidene)-2-(p-tolyl)thiazol-4(5H)-one (32)
1
A brown solid (64 mg, 36%). mp 240–242 °C; HPLC purity: 18.45 min, 93.5%. H
NMR (600 MHz, DMSO-d6) δ 8.12 (d, J = 7.7 Hz, 2H), 8.01 (s, 1H), 7.80 (d, J = 8.1
13
Hz, 2H), 7.76 (d, J = 8.1 Hz, 2H), 7.50 (d, J = 7.7 Hz, 2H), 2.46 (s, 3H). C NMR
(150 MHz, DMSO-d6) δ 186.3, 182.1, 147.0, 136.1, 132.7, 132.5, 132.3, 130.3, 128.7,
128.4, 127.2, 124.9, 21.5. HRMS (ESI+): calcd for [M+H]+ C17H13BrNOS+ m/z
357.9896; found 357.9893.
(Z)-5-(2-Fluorobenzylidene)-2-(p-tolyl)thiazol-4(5H)-one (33)
1
A yellow solid (45 mg, 30%). mp 192–194 °C; HPLC purity: 18.61 min, 93.7%. H
NMR (600 MHz, CDCl3) δ 8.29 (s, 1H), 8.11 (d, J = 7.8 Hz, 2H), 7.74 (t, J = 7.2 Hz,
1H), 7.49-7.43 (m, 1H), 7.37 (d, J = 7.8 Hz, 2H), 7.30 (t, J = 7.2 Hz, 1H), 7.21-7.16
(m, 1H), 2.48 (s, 3H). 13C NMR (150 MHz, CDCl3) δ 187.2, 183.1, 162.0 (d, J =
256.3 Hz), 147.0, 132.9 (d, J = 8.8 Hz), 130.2, 129.7 (d, J = 6.5 Hz), 129.4, 129.2,
129.1, 128.6, 124.8 (d, J = 4.7 Hz), 122.5 (d, J = 11.7 Hz), 116.4 (d, J = 21.7 Hz),
22.0. HRMS (ESI+): calcd for [M+H]+ C17H13FNOS+ m/z 298.0696; found 298.0690.
(Z)-5-(4-Fluorobenzylidene)-2-(p-tolyl)thiazol-4(5H)-one (34)
1
A yellow solid (42 mg, 28%). mp 226–228 °C; HPLC purity: 17.43 min, 93.8%. H
15

NMR (600 MHz, CDCl3) δ 8.11 (d, J = 8.2 Hz, 2H), 8.01 (s, 1H), 7.68 (dd, J = 5.3,
13
8.5 Hz, 2H), 7.37 (d, J = 8.2 Hz, 2H), 7.20 (t, J = 8.5, 8.5 Hz, 2H), 2.48 (s, 3H). C
NMR (150 MHz, CDCl3) δ 186.9, 183.3, 164.1 (d, J = 255 Hz), 146.8, 136.7, 132.8 (d,
J =8.7 Hz), 130.4 (d, J = 3.3 Hz), 130.2, 129.3, 129.1, 126.4 (d, J = 2.6 Hz), 116.7 (d,
J = 22.1 Hz), 22.0. HRMS (ESI+): calcd for [M+H]+ C17H13FNOS+ m/z 298.0696;
found 298.0689.
(Z)-5-(2,4-Difluorobenzylidene)-2-(p-tolyl)thiazol-4(5H)-one (35)
1
A yellow solid (55 mg, 35%). mp 196–198 °C; HPLC purity: 17.57 min, 93.6%. H
NMR (600 MHz, CDCl3) δ 8.20 (s, 1H), 8.10 (d, J = 8.2 Hz, 2H), 7.75-7.71 (m, 1H),
7.37 (d, J = 8.2 Hz, 2H), 7.06-7.03 (m, 1H), 6.96-6.92 (m, 1H), 2.48 (s, 3H). 13C
NMR (150 MHz, CDCl3) δ 186.9, 183.0, 163.5 (dd, J = 12.2, 255 Hz), 162.5 (dd, J =
12.0, 255 Hz), 147.2, 130.6 (dd, J = 3.0, 10.0 Hz), 130.2, 129.2, 129.2, 128.4 (dd, J =
1.7, 6.0 Hz), 128.2, 119.1 (dd, J = 3.9, 12.0, Hz), 112.3 (dd, J = 3.7, 21.8 Hz), 105.1
(dd, J = 25.5, 25.5 Hz), 22.2. HRMS (ESI+): calcd for [M+H]+ C17H12F2NOS+ m/z
316.0602; found 316.0607.
(Z)-5-(4-(Benzyloxy)benzylidene)-2-(p-tolyl)thiazol-4(5H)-one (36)
1
An orange solid (66 mg, 34%). mp 182–184 °C; HPLC purity: 17.71 min, 95.6%. H
NMR (600 MHz, CDCl3) δ 8.11 (d, J = 8.2 Hz, 2H), 8.01 (s, 1H), 7.64 (d, J =8.3 Hz,
2H), 7.45-7.40 (m, 4H), 7.41 (d, J = 7.8 Hz, 2H), 7.37-7.35 (m, 3H), 7.09 (d, J = 8.8
Hz, 2H), 5.15 (s, 2H), 2.47 (s, 3H). 13C NMR (150 MHz, CDCl3) δ 186.6, 183.8,
161.2, 146.5, 138.2, 136.2, 132.9, 130.1, 129.5, 129.0, 128.9, 128.5, 127.6, 126.9,
124.0, 115.8, 70.4, 22.1. HRMS (ESI+): calcd for [M+H]+ C24H20NO2S+ m/z 386.1209;
found 386.1211.
(Z)-Methyl 4-((4-oxo-2-(p-tolyl)thiazol-5(4H)-ylidene)methyl)benzoate (37)
A yellow-orange solid (49 mg, 29%). mp 212–214 °C; HPLC purity: 18.94 min,
1
95.1%. H NMR (600 MHz, CDCl3) δ 8.14 (d, J = 8.2 Hz, 2H), 8.11 (d, J = 8.2 Hz,
2H), 8.02 (s, 1H), 7.71 (d, J = 8.2 Hz, 2H), 7.37 (d, J = 8.2 Hz, 2H), 3.95 (s, 3H), 2.48
(s, 3H). 13C NMR (150 MHz, CDCl3) δ 187.3, 183.2, 166.3, 147.3, 138.1, 136.3,
131.7, 130.4, 130.4, 130.2, 129.2, 129.1, 129.0, 52.6, 22.2. HRMS (ESI+): calcd for
[M+H]+ C19H16NOS+ m/z 338.0845; found 338.0847.
16

In vitro Enzymatic Assay
The in vitro enzymatic assay against PRMT5 was carried out by Shanghai
ChemPartner Co. (998 Halei Road, Pudong New District, Shanghai 201203, China).
All of the inhibitory rate data were obtained by performing triplicate tests and showed
as the mean value. IC50 values were tested with eight different concentrations (from
100 μM, times dilution) of each compound and showed as the mean ± SD of three
replicates. The following materials were purchased: PRMT5/MEP50 (BPS, Cat. No.
51045); PRMT1 (BPS, Cat. No. 51041); [3H]-SAM (PerkinElmer, Lot. No. 2146246);
SAM (Sigma, Cat. No. A7007); SAH (Sigma, Cat. No. A9384-25MG); 384-well plate
(Perkin Elmer, Cat. No. 6007299); Anti-methyl-H4R3 AlphaLISA acceptor beads,
streptavidin-tagged donor beads (Perkin Elmer); 1 x Epigenetics Buffer (Perkin
Elmer). SAH was used as the reference compound. All tested compounds were
dissolved into 20 mM stock in 100% DMSO. AlphaLISA and radioactive methylation
assays were used to
test the inhibitory activities against PRMT5 and PRMT1,
respectively, following the method we previously reported.15 The following protocol
was used to test the inhibitory activity against PRMT5. Briefly, the biotinylated
unmethylated
histone
H4
peptide
(the
23-amino
acid
sequence:
SGRGKGGKGLGKGGAKRHRKVLRGGK(biotin)–NH2) substrate at arginine 3
(unmeH4R3) substrate stock and the SAM stock was diluted using Assay buffer
(10mM Tris, PH 8.0, 1mM DTT, 0.01% Tween-20) with final concentrations of 50
nM and 64 μM respectively in the reaction well. The PRMT5/MEP50 enzyme was
diluted at a ratio of 1 : 10 in Assay buffer before being used. 100 nL of compound
solutions were transferred to each well of the assay plate. Next, 5 μL of
unmeH4R3:SAM mixture and 5 μL of diluted enzyme stock (final concentration of
0.25 nM) were mixed in each well and incubated at room temperature (RT) for 60 min.
10 μL of the assay buffer was used to serve as positive ‘‘maximum signal’’ control (no
compound, with PRMT5 enzyme added), while “background signal’’ control was with
no compound and no enzyme added. After that, 7.5 μL anti-H4R3me2 acceptor beads
(1 : 300 dilution of the 5 mg/ml stock in 1 x Epigenetics buffer) and 7.5 μL
17

streptavidin-coated donor bead beads (1 : 300 dilution of the 5 mg/mL stock in 1x
Epigenetics buffer) were successively added to each well of the assay plates with an
interval of 60 min. The plates were incubated at RT for 30 min in the dark followed
by Alpha signal reading on an EnSpire reader. GraphPad Prism 5.0 was used to
process data.
The following protocol was used to test the inhibitory activity of compound 5 against
PRMT1. The enzyme (PRMT1), substrate (H4R3 S1ac), [3H]-SAM and Stop Mix
(SAM) solution were prepared in assay buffer with the final concentration of 0.5 nM,
0.1 μM, 0.25 μM and 0.13 mM, respectively. Compounds were added to the assay
plate by Echo. 15 µL Enzyme solution was added to per well of prepared compound
stock plates, followed by addition of 10 µL peptide and [3H]-SAM to per well of
prepared compound stock plate using Multidrop. The plates were then covered and
incubated for 60 minutes at room temperature. The reaction was finally stopped by
addition of 5 µL Stop Mix. 25 µL solution from per well of the compound stock plate
was transferred to Flashplate using Platemate, and the Flashplate was covered and
then incubated for another 60 min at room temperature. The Flashplate was washed
with dH2O + 0.1% Tween using the BioTek plate washer and then read plate in
TopCount using program 3H-flashplate.
Molecular dynamics simulation
100 ns MD simulations were performed on 5-PRMT5:MEP50 and
19-PRMT5:MEP50 complex models obtained by molecular docking. H++ program26
was used to predict the protonation states of ionizable residues of each model. A
periodic box of transferable intermolecular potential 3P water molecules that extended
10.0 Å from the protein atoms was performed on each complex model. In order to
neutralize the simulation system, counterions were added. AMBER 14.0 package27
was used to perform MD simulations. Each model was done with isothermal-isobaric
(NPT) ensemble and periodic boundary conditions. The force field for protein and
small molecules were Amber14SB and the general Amber force field (GAFF)28
respectively. Antechamber29 was used to derive charges and force field parameters of
small molecules that were not existent in GAFF. SHAKE algorithm30 and
18

particle-mesh Ewald method31 were respectively used to constrain small bonds
involving hydrogen atoms and to calculate the electrostatic interactions. The
non-bonded pairs were updated every 25 steps while the non-bonded cutoff was set to
10 Å. Each MD simulation was coupled to a 300 K thermal bath at 1 bar pressure by
using the algorithm of Berendsen et al.32
MM/PBSA calculations.
After the trajectory of each complex model was well equilibrated, binding free energy
was calculated using the MM/PBSA method encoded in the AMBER 14.0 program.
2000 snapshots from each model trajectory were extracted every 50 ps, and the
MM/PBSA calculation was performed on each snapshot using the MMPBSA.py.MPI
module.
In vitro anti-proliferative assay on MV4-11 cell line.
Cell viability was measured by the method described previously.15 Briefly, MV4-11
cells were plated into 24-well plates with 1 mL of RPMI1640 medium (Invitrogen)
containing 10% fetal bovine serum and 1% penicillin/streptomycin at a density of
0.5–1 x 105 cells per well. The cells were incubated with tested compounds at
different concentrations or DMSO at 37 °C in an incubator with a supply of 5% CO2
for four, eight, and 12 days. Viable cell number was determined using the Cell
Titer-Glo Luminescent Cell Viability assays (Promega), and luminescence was
recorded using an EnVision Multilabel Plate Readers (PerkinElmer) according to the
manufacturer’s protocol EC50 values were determined from at least three independent
tests and calculated from the inhibition curves.
Western Blotting
Total cell lysates were separated by 4%−1 % SDS-polyacrylamide gels and
transferred to nitrocellulose membranes. Then blocking buffer (5% nonfat milk in
PBST) and primary antibodies (SDMA, CST, 13222s; GAPDH, CST, 5174; both
prepared in a 1:3 dilution of blocking buffer) were subsequently used to block and
19

incubate the blots for one hour at room temperature and overnight at 4 °C,
respectively. After being washed three times with PBST, the blots were incubated with
1:10000 dilution of donkey anti-rabbit secondary antibody (HRP conjugated) for one
hour. Finally, the bands were detected in the ChemiScope3400 imaging system using
ECL substrate (Millipore).
Flow Cytometric Analysis
MV4-11 cells were plated in 6-well plates and treated with tested compounds (5 and
19) or DMSO (control). For cell cycle analysis, cells harvested after 24 hours were
re-suspended in 70% ethanol overnight at 4 °C for fixation. After washed with PBS,
samples were incubated with Propidium Iodide/RNase Staining Buffer (BD
Pharmingen) for 30 minutes at room temperature. For cell apoptosis analysis, cells
harvested after 48 hours were measured using AnnexinV-FITC Apoptosis Detection
Kit (Vazyme Biotech) according to the manufacturer’s protocol. Samples were
detected by BD FACSCalibur (BD Pharmingen) and data were analyzed using FlowJo
V7.6.1.
Conclusions
In conclusion, 11 novel PRMT5 inhibitors were identified with IC50 values ranging
from 0.77 to 23 μM by docking-based virtual screening combined with biological
evaluation. Compared with the previously reported PRMT5 inhibitors, the hits
identified in this work contained a new scaffold and could be used for further
optimization to improve their activities. From the proposed binding mode, the top two
active hits (5 and 19) had hydrophobic, hydrogen bond, salt bridge and cation-π
interactions with PRMT5. MD simulation results indicated that the order of the
binding free energy was in well agreement with the experimental activity. In addition,
the anti-proliferative effect and the cell-level target engagement of 5 and 19 were
confirmed in MV4-11 cells. Further investigations uncovered the G1 cell-cycle arrest
and inducement of cell apoptosis to be the mechanism of cellular activity. The hits
discovered in this study has provided a new chemical template for further hit-to-lead
20

optimization and lay a foundation for further development of therapeutic candidates
for cancer.
Abbreviations
PRMT, protein arginine methyltransferase; SAM, sadenosylmethionine; MEP50,
methylosome protein 50; MD, molecular dynamics; MM/PBSA, Molecular
mechanics/Poisson-Boltzmann surface area; SAR, structure-activity relationship.
Acknowledgements
This work was supported by the Shandong Provincial Natural Science Foundation
[Nos. ZR2017BH038, JQ201721], National Natural Science Foundation of China
[Nos. 21672082, 21472208, 81625022, and 81430084], Shandong Key Development
Project [No. 2016GSF201209], the Young Taishan Scholars Program [No.
tsqn20161037], Shandong Talents Team Cultivation Plan of University Preponderant
Discipline [No. 10027], and China Postdoctoral Science Foundation [2016M590391].
Reference
1.
2.
3.
V. Karkhanis, Y. J. Hu, R. A. Baiocchi, A. N. Imbalzano and S. Sif, Trends
Biochem. Sci., 2011, 36, 633-641.
E. S. Burgos, C. Wilczek, T. Onikubo, J. B. Bonanno, J. Jansong, U. Reimer
and D. Shechter, J. Biol. Chem., 2015, 290, 9674-9689.
Q. Zhao, G. Rank, Y. T. Tan, H. Li, R. L. Moritz, R. J. Simpson, L. Cerruti, D.
J. Curtis, D. J. Patel, C. D. Allis, J. M. Cunningham and S. M. Jane, Nat.
Struct. Mo.l Biol., 2009, 16, 304-311.
4.
5.
S. Pal, S. N. Vishwanath, H. Erdjument-Bromage, P. Tempst and S. Sif, Mol.
Cell Biol., 2004, 24, 9630-9645.
E. Chan-Penebre, K. G. Kuplast, C. R. Majer, P. A. Boriack-Sjodin, T. J. Wigle,
L. D. Johnston, N. Rioux, M. J. Munchhof, L. Jin, S. L. Jacques, K. A. West, T.
Lingaraj, K. Stickland, S. A. Ribich, A. Raimondi, M. P. Scott, N. J. Waters, R.
M. Pollock, J. J. Smith, O. Barbash, M. Pappalardi, T. F. Ho, K. Nurse, K. P.
Oza, K. T. Gallagher, R. Kruger, M. P. Moyer, R. A. Copeland, R. Chesworth
and K. W. Duncan, Nat. Chem. Biol., 2015, 11, 432-437.
6.
7.
X. Sheng and Z. Wang, BMC Cancer, 2016, 16, 567.
F. Yan, L. Alinari, M. E. Lustberg, L. K. Martin, H. M. Cordero-Nieves, Y.
Banasavadi-Siddegowda, S. Virk, J. Barnholtz-Sloan, E. H. Bell, J. Wojton, N.
K. Jacob, A. Chakravarti, M. O. Nowicki, X. Wu, R. Lapalombella, J. Datta, B.
21

Yu, K. Gordon, A. Haseley, J. T. Patton, P. L. Smith, J. Ryu, X. Zhang, X. Mo,
G. Marcucci, G. Nuovo, C. H. Kwon, J. C. Byrd, E. A. Chiocca, C. Li, S. Sif,
S. Jacob, S. Lawler, B. Kaur and R. A. Baiocchi, Cancer Res, 2014, 74,
1752-1765.
8.
9.
M. A. Powers, M. M. Fay, R. E. Factor, A. L. Welm and K. S. Ullman, Cancer
Res., 2011, 71, 5579-5587.
J. Chung, V. Karkhanis, S. Tae, F. Yan, P. Smith, L. W. Ayers, C. Agostinelli, S.
Pileri, G. V. Denis, R. A. Baiocchi and S. Sif, J. Biol. Chem., 2013, 288,
35534-35547.
10.
11.
L. Wang, S. Pal and S. Sif, Mol. Cell. Biol., 2008, 28, 6262-6277.
S. Pal, R. A. Baiocchi, J. C. Byrd, M. R. Grever, S. T. Jacob and S. Sif, EMBO
J., 2007, 26, 3558-3569.
12. F. Ye, W. Zhang, X. Ye, J. Jin, Z. Lv and C. Luo, J. Chem. Inf. Model., 2018, 58,
1066-1073.
13.
S. Antonysamy, Z. Bonday, R. M. Campbell, B. Doyle, Z. Druzina, T. Gheyi,
B. Han, L. N. Jungheim, Y. W. Qian, C. Rauch, M. Russell, J. M. Sauder, S. R.
Wasserman, K. Weichert, F. S. Willard, A. P. Zhang and S. Emtage, P. Natl.
Acad. Sci. USA., 2012, 109, 17960-17965.
14.
Y. Ye, B. Zhang, R. Mao, C. Zhang, Y. Wang, J. Xing, Y. C. Liu, X. Luo, H.
Ding, Y. Yang, B. Zhou, H. Jiang, K. Chen, C. Luo and M. Zheng, Org.
Biomol. Chem., 2017, 15, 3648-3661.
15.
16.
17.
18.
R. Mao, J. Shao, K. Zhu, Y. Zhang, H. Ding, C. Zhang, Z. Shi, H. Jiang, D.
Sun, W. Duan and C. Luo, J. Med. Chem., 2017, 60, 6289-6304.
S. Ji, S. Ma, W. J. Wang, S. Z. Huang, T. Q. Wang, R. Xiang, Y. G. Hu, Q.
Chen, L. L. Li and S. Y. Yang, Chem. Biol. Drug Des., 2017, 89, 585-598.
G. B. Gonsalvez, L. Tian, J. K. Ospina, F. M. Boisvert, A. I. Lamond and A. G.
Matera, J. Cell Biol., 2007, 178, 733-740.
W. J. Friesen, S. Paushkin, A. Wyce, S. Massenet, G. S. Pesiridis, G. Van
Duyne, J. Rappsilber, M. Mann and G. Dreyfuss, Mol. Cell Biol., 2001, 21,
8289-8300.
19.
20.
G. Meister, C. Eggert, D. Buhler, H. Brahms, C. Kambach and U. Fischer,
Curr. Biol., 2001, 11, 1990-1994.
K. W. Duncan, N. Rioux, P. A. Boriack-Sjodin, M. J. Munchhof, L. A. Reiter,
C. R. Majer, L. Jin, L. D. Johnston, E. Chan-Penebre, K. G. Kuplast, M. P.
Scott, R. M. Pollock, N. J. Waters, J. J. Smith, M. P. Moyer, R. A. Copeland
and R. Chesworth, ACS Med. Chem. Lett., 2016, 7, 162-166.
21.
22.
K. Mavrakis, E. R. McDonald, M. R. Schlabach, E. Billy, G. R. Hoffman, A.
deWeck, D. A. Ruddy, K. Venkatesan, G. McAllister, R. deBeaumont, S. Ho, Y.
Liu, Y. Yan-Neale, G. Z. Yang, F. Lin, H. Yin, H. Gao, D. R. Kipp, S. P. Zhao,
J. T. McNamara, E. R. Sprague, Y. S. Cho, J. Gu, K. Crawford, V. Capka, K.
Hurov, J. A. Porter, J. Tallarico, C. Mickanin, E. Lees, R. Pagliarini, N. Keen,
T. Schmelzle, F. Hofmann, F. Stegmeier and W. R. Sellers, Science, 2016, 351,
1208-1213.
B. Hofmann, S. Barzen, C. B. Rodl, A. Kiehl, J. Borig, A. Zivkovic, H. Stark,
22

G. Schneider and D. Steinhilber, J. Med. Chem., 2011, 54, 1943-1947.
Y. K. Banasavadi-Siddegowda, L. Russell, E. Frair, V. A. Karkhanis, T.
Relation, J. Y. Yoo, J. Zhang, S. Sif, J. Imitola, R. Baiocchi and B. Kaur,
Oncogene, 2017, 36, 263-274.
23.
24.
25.
T. A. Halgren, R. B. Murphy, R. A. Friesner, H. S. Beard, L. L. Frye, W. T.
Pollard and J. L. Banks, J. Med. Chem., 2004, 47, 1750-1759.
R. A. Friesner, J. L. Banks, R. B. Murphy, T. A. Halgren, J. J. Klicic, D. T.
Mainz, M. P. Repasky, E. H. Knoll, M. Shelley, J. K. Perry, D. E. Shaw, P.
Francis and P. S. Shenkin, J. Med. Chem., 2004, 47, 1739-1749.
J. C. Gordon, J. B. Myers, T. Folta, V. Shoja, L. S. Heath and A. Onufriev,
Nucleic Acids Res., 2005, 33, W368-371.
D. A. Case, T. E. Cheatham, 3rd, T. Darden, H. Gohlke, R. Luo, K. M. Merz,
Jr., A. Onufriev, C. Simmerling, B. Wang and R. J. Woods, J. Comput. Chem.,
2005, 26, 1668-1688.
26.
27.
28.
29.
30.
J. Wang, R. M. Wolf, J. W. Caldwell, P. A. Kollman and D. A. Case, J. Comput.
Chem., 2004, 25, 1157-1174.
J. Wang, W. Wang, P. A. Kollman and D. A. Case, J. Mol. Graph. Model.,
2006, 25, 247-260.
J. P. Ryckaert, G. Ciccotti, H. J. C. Berendsen, J. Comput. Phys., 1977, 23,
327-341.
31.
32.
T. Darden, Y. Darrin, L. Pedersen, J. Chem. Phys. 1993, 98, 10089-10092.
H. J. C. Berendsen, J. P. M. Postma, W. F. V. Gunsteren, A. Dinola, J. R. Haak,
J. Chem. Phys., 1984, 81, 3684-3690.
23

Table 1 Inhibitory activity of P16 and its analogues (1 to 44) against PRMT5a
R1
R2
Inhibition rate (%)
62
IC50 (μM)
Source
Specs
Compd. No.
P16
Me
13.9 ± 0.85
Me
Me
Me
Me
5
9
N.D.
N.D.
N.D.
N.D.
Synthesis
Synthesis
Synthesis
Synthesis
1
2
3
4
12
33
Me
Me
Me
Me
Me
Me
Me
82
24
15
7
0.77 ± 0.04 Synthesis
5
6
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
Synthesis
Synthesis
Synthesis
Specs
7
8
6
9
20
18
Synthesis
Synthesis
10
11
24

Me
30
N.D.
Specs
12
Me
Me
Me
27
56
80
N.D.
Specs
Specs
Specs
13
14
15
16 ± 0.71
12 ± 1.41
Me
Me
Me
44
28
33
23 ± 2.12
N.D.
Specs
Specs
Specs
16
17
18
N.D.
Me
98
6.6 ± 0.50
Specs
19
Me
Me
42
0
18 ± 0.85
N.D.
Specs
Specs
20
21
Me
89
16 ± 0.14
Specs
22
Me
Me
30
4
N.D.
N.D.
Synthesis
23
24
Synthesis
25

Me
0
N.D.
Synthesis
25
Me
Me
Me
14
4
N.D.
N.D.
N.D.
Synthesis
Synthesis
Synthesis
26
27
28
14
Me
Me
Me
Me
Me
Me
Me
Me
0
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
Synthesis
Synthesis
Synthesis
Synthesis
Synthesis
Synthesis
Synthesis
Synthesis
29
30
31
32
33
34
35
36
30
14
0
25
2
17
17
Me
H
0
26
8
N.D.
N.D.
N.D.
Synthesis
Specs
37
38
39
H
Specs
H
2
N.D.
Specs
40
26

H
H
H
48
65
23
11 ± 0.35
15 ± 0.64
N.D.
Specs
Specs
Specs
41
42
43
H
93
17± 0.28
Specs
44
a
The initial inhibition rates were tested at 50 μM, and only compounds with >40%
inhibition rates were selected for IC50 measurements. N.D. = not determined. Data
shown are mean ± SD of three replicates.
27

Fig. 1 Previously reported PRMT5 inhibitors of six scaffolds.
Fig. 2 Virtual screening procedures and activity assay results for PRMT5 in vitro. (A)
Workflow of virtual screening. (B) Structure and inhibitory activity of P16 against
PRMT5 at enzymatic level.
28

Fig. 3 Chemical structures and inhibitory activities against PRMT5 of 5 and 19 and
their predicted binding modes with PRMT5. (A) Structures and inhibition curves of
compounds 5 and 19 against PRMT5. (B) Binding mode analysis of 5 and 19 with
PRMT5. PRMT5, MEP50 and H4 peptide are shown in cartoon diagram in yellow,
orange, and cyan color, respectively, and 5 and 19 are represented as green and
salmon sticks, respectively. (C) A close-up view of the binding modes of 5 and 19.
Arginine 3 of H4 peptide is shown as cyan sticks.
Fig. 4 The detailed interactions of 5 and 19 with PRMT5 obtained by molecular
29