GSK3235025

EPZ015666, a selective protein arginine methyltransferase 5 (PRMT5) inhibitor with an antitumour effect in retinoblastoma

Xing Liu, JianZhong He, Longbing Mao, Yanyan Zhang, WenWen Cui, Sujuan Duan, Alan Jiang, Yang Gao, Yi Sang, Guofu Huang

PII: S0014-4835(20)30544-3
DOI: https://doi.org/10.1016/j.exer.2020.108286 Reference: YEXER 108286

To appear in: Experimental Eye Research

Received Date: 10 December 2019
Revised Date: 19 September 2020
Accepted Date: 29 September 2020

Please cite this article as: Liu, X., He, J., Mao, L., Zhang, Y., Cui, W., Duan, S., Jiang, A., Gao, Y., Sang, Y., Huang, G., EPZ015666, a selective protein arginine methyltransferase 5 (PRMT5) inhibitor with an antitumour effect in retinoblastoma, Experimental Eye Research (2020), doi: https://doi.org/10.1016/ j.exer.2020.108286.

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. 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.
© 2020 Published by Elsevier Ltd.

⦁ EPZ015666, a selective protein arginine methyltransferase 5 (PRMT5)
⦁ inhibitor with an antitumour effect in retinoblastoma

⦁ Xing Liua,1, JianZhong Heb,1, Longbing Maoa, Yanyan Zhanga, WenWen Cuia, Sujuan Duana, Alan
⦁ Jiangc, Yang Gaoa, Yi Sangc, and Guofu Huanga,c,*
5
6
⦁ a Department of Ophthalmology, The Third Affiliated Hospital of Nanchang University, 128
⦁ Xiangshan Northern Road, Nanchang City, 330008 Jiangxi Province, People’s Republic of China.
⦁ b Department of Ophthalmology, The People’s Hospital of Pingxiang City, Pingxiang City, 337055
⦁ Jiangxi Province, People’s Republic of China.
⦁ c Jiangxi Key Laboratory of Cancer Metastasis and Precision Treatment, The Third Affiliated
⦁ Hospital of Nanchang University, 128 Xiangshan Northern Road, Nanchang City, 330008 Jiangxi
⦁ Province, People’s Republic of China.
⦁ 1 These authors contributed equally to the manuscript.
15
⦁ *Corresponding author:
⦁ Professor Guofu Huang,
⦁ E-mail: ⦁ [email protected]
⦁ Department of Ophthalmology, The Third Affiliated Hospital of Nanchang University, 128 Xiangshan
⦁ Northern Road, Nanchang, 330008 Jiangxi People’s Republic of China.
21
22
⦁ Abstract
⦁ Retinoblastoma (RB) is the most common intraocular malignant tumour in infants, and
⦁ chemotherapy has been the primary therapy method in recent years. PRMT5 is an important
⦁ member of the protein arginine methyltransferase family, which plays an important role in various
⦁ tumours. Our study showed that PRMT5 was overexpressed in retinoblastoma and played an
⦁ important role in retinoblastoma cell growth. EPZ015666 is a novel PRMT5 inhibitor, and we found
⦁ that it inhibited retinoblastoma cell proliferation and led to cell cycle arrest at the G1 phase. At the
⦁ same time, EPZ015666 regulated cell cycle related protein (P53, P21, P27, CDK2) expression. In
⦁ brief, our study showed that PRMT5 promoted retinoblastoma growth, the PRMT5 inhibitor
⦁ EPZ015666 inhibited retinoblastoma in vitro by regulating P53-P21/P27-CDK2 signalling pathways
⦁ and slowed retinoblastoma growth in a xenograft model.
⦁ Keywords
⦁ Retinoblastoma; PRMT5; EPZ015666.
36
37 Specifications Table
38
Subject Ophthalmology
Specific subject area The research provided a novel antitumour drug for retinoblastoma.

Type of data Figure
How data were acquired Microscope, Flow Cytometry, MTT,real-time PCR
Instruments: CFX96 Real-Time System. BIO-Rad; FACS Calibur flow Cytometry; BD.
Data format Analysis
Parameters for data collection Cells were collected for experiments when they were in the logarithmic growth phase. Prior to analysis, the same number of cells were treated in the same environment and for the same period of time.
Description of data collection MTT assay was used to access cells viability, flow cytometry analysis was used for cell cycle, real-time PCR and Western blotting were used to assess mRNA and protein levels, respectively. All of the experiments were repeated three times.
Data source location Institution: The Third Affiliated Hospital of Nanchang University City/Town/Region: Nanchang, Jiangxi
Country:China
Latitude and longitude (and GPS coordinates) for collected samples/data: North latitude N28°40′56.68″ East longitude E115°53′9.22″
Data accessibility With the article
1
2
3 Value of the Data
4
⦁ 1. There is an urgent need for novel strategies or drugs for retinoblastoma.
⦁ 2. Protein arginine methyltransferase 5 plays an important role in retinoblastoma. EPZ015666, a
⦁ selective PRMT5 inhibitor, exerts antitumour effects in retinoblastoma.
⦁ 3. EPZ015666 is a potential strategy for retinoblastoma patients.
9
10 Data Description
11
⦁ Fig. 1. PRMT5 was overexpressed in RB. (A, B) The mRNA and protein levels of PRMT5 in
⦁ retinoblastoma cell lines and retinal pigment epithelium cells (ARPE-19) as controls. (C, D)
⦁ Immunohistochemistry shows the expression level of PRMT5 in human retinoblastoma tissues and
⦁ adjacent tissues using two different PRMT5 monoclonal antibodies. (C) representative images. (D)
⦁ IHC score in retinoblastoma tissues and adjacent tissues. (E) The protein level of PRMT5 was
⦁ detected by Western blotting. N represents adjacent tissues and T represents retinoblastoma
⦁ tissues. β-actin is a loading control. (*p<0.01 and **p<0.001)
19
⦁ Fig. 2. The effects of PRMT5 on retinoblastoma cells. (A) Y79 cells stably overexpressing PRMT5.
⦁ (B) MTT assay showed the cell viability of PRMT5-overexpressing cells. (C) Colony formation assay
⦁ showed the number of colonies after overexpressing PRMT5. (D) Flow cytometry analysis of the Y79
⦁ cell cycle. (E) Knockdown of PRMT5 by shRNA in WERI-Rb-1 cells. (F) MTT assay showed the cell
⦁ viability after knocking down PRMT5. (G) Colony formation assay showed the number of colonies

1 after knocking down PRMT5. (H) Flow cytometry analysis of the WERI-Rb-1 cell cycle after knocking
2 down PRMT5. (*p<0.01)
⦁ Fig. 3. EPZ015666 inhibited RB cell growth. (A, B) Y79 and WERI-Rb-1 activity were inhibited by
⦁ EPZ015666 in a time and dose-dependent manner. (C, D) Colony assay showed that the number of
⦁ colonies decreased after RB cells were treated with EPZ015666. (E, F) Flow cytometry analysis
⦁ indicated that EPZ015666 resulted in G1 arrest in RB cells (*p<0.01 and **p<0.001).
⦁ Fig. 4. EPZ015666 regulated cell cycle-related factor expression. Y79 and WERI-Rb-1 cells were
⦁ treated with various concentrations of EPZ015666 for three days. Western blot analysis showed that
⦁ the expression of P53, P27, and P21 was upregulated, while the expression of CDK2 was
⦁ downregulated after RB cells were treated with EPZ015666.
⦁ Fig. 5. EPZ015666 inhibited tumour growth in Y79 cell xenografts. The antitumour effect induced
⦁ after oral administration of EPZ015666 compared to that oral administration of vehicle in xenografted
⦁ mice. (A) Mice images after treatment with vehicle or EPZ015666 (150 mg/kg) for 21 days. (B) The
⦁ tumour sizes during the treatment period. (C) Tumour images after treatment with vehicle or
⦁ EPZ015666 for 21 days. (D) Tumour weights after treatment with vehicle or EPZ015666 for 21 days.
⦁ (E) The body weights of the two groups of mice during the treatment period (*p<0.01).
17
18 Experimental Design, Materials, and Methods
19

⦁ 1. Introduction
⦁ Retinoblastoma(RB) is the most common intraocular malignant tumour in infants, and has a
⦁ worldwide incidence rate of 1:15,000~1:20,000 live births (ID et al., 2018). Although it has a low
⦁ incidence rate, RB has received much attention from clinicians due to its effects on vision and life. At
⦁ present, enucleation, intravenous chemotherapy (IVC), and intra-ophthalmic artery chemotherapy
⦁ (IAC) are the most commonly used treatments. Enucleation is a disruptive surgery that requires
⦁ general anaesthesia and has some effects on the patient’s mental health to some degree, so it is
⦁ usually the choice only in case of very large or advanced tumours where it is required to save the
⦁ patient’s life (Y et al., 2018; ID et al., 2018). IVC has been the most common therapy method in
⦁ recent years, with the common chemotherapy drugs used being carboplatin (CBP), vincristine
⦁ (VCR), and etoposide (CL et al., 2013). A large number of reports indicate that CBP, VCR, and
⦁ etoposide have serious duplicate effects on patients (MA et al., 2019a; D et al., 2016; AS et al.,
⦁ 2019; MA et al., 2019b). IAC is a novel therapy method, and the common IAC drugs are melphalan,
⦁ topotecan, and CBP. However, the drugs can disrupt the function of the entire retina layer (AB et al.,
⦁ 2019; D et al., 2016). Moreover, IAC can lead to vascular events or metastatic disease, and it
⦁ requires advanced technology. Thus, new therapy strategies or targets are urgently needed.
⦁ Protein arginine methylation is a ubiquitous post-translational modification and is promoted by
⦁ protein arginine methyltransferases (PRMTs). PRMT5 is a primary member of the PRMT family (N et
⦁ al., 2015), that plays an important role in the cellular processes involved in transcriptional repression,
⦁ RNA splicing, and signal transduction. While recent studies indicated that PRMT5 is also related to
⦁ cell growth and differentiation (D et al., 2017; S et al., 2004; TN et al., 2012), increasing numbers of
⦁ reports have revealed PRMT5 overexpression in various tumours, noting that it promotes the
⦁ development of malignancy in cancers such as, hepatocellular carcinoma (JY et al., 2018; Z et al.,
43 2018a; Z et al., 2018b; B et al., 2015), breast cancer (Y et al., 2017; Z et al., 2018c; H et al., 2019),

⦁ gastric cancer (B et al., 2017), lung cancer (P et al., 2018; S et al., 2019), and lymphoma (J et al.,
⦁ 2013). These studies also note that knockdown of PRMT5 or treatment with a PRMT5 inhibitor might
⦁ slow or stop tumour growth.
⦁ EPZ015666 is a novel selective PRMT5 inhibitor that has been tested for its antitumour role via
⦁ inhibition of PRMT5 in various cancers (S et al., 2018; SV et al., 2018; G et al., 2018; B et al., 2019).
⦁ In glioblastoma, EPZ015666 exerted synergistic antitumour effects in vitro and in a xenograft mouse
⦁ model when used in combination with torkinib. The expression of PRMT5 is upregulated in bladder
⦁ cancer cell lines, and EPZ015666 inhibited bladder cancer cell proliferation and led to apoptosis via
⦁ suppression of NF-κB activation (G et al., 2018). Oral dosing with EPZ015666 resulted in dose-
⦁ dependent antitumour activity in mantle cell lymphoma xenograft models (K et al., 2019b). In
⦁ addition, EPZ015666 has been evaluated in phase I clinical trials for the treatment of several
⦁ haematological malignancies, including lymphoma.(X et al., 2019; E et al., 2015).
⦁ However, the effects of PRMT5 and EPZ015666 on retinoblastoma have not been examined to
⦁ date. Our research focused on studying the expression and role of PRMT5 in RB growth and aimed
⦁ to determine whether the PRMT5 inhibitor EPZ015666 can inhibit the progression of retinoblastoma.
⦁ 2. Materials and methods
⦁ 2.1. Cell lines
⦁ Human APRE-19 cell, 293T cell and retinoblastoma cell lines Y79, and WERI-Rb-1 were obtained
⦁ from the Institute of Biochemistry and Cell Biology (Shanghai, China). The APRE-19 cells and 293T
⦁ cells were cultured with DMEM (Gibco, Invitrogen, Waltham, MA, USA) containing 10% foetal bovine
⦁ serum (Gibco, Invitrogen) and 1% penicillin/streptomycin (Invitrogen, 15140) in a 37 °C humidified
⦁ incubator with 5% CO2. Y79 and WERI-Rb-1 cells were cultured with RPMI-1640 (Gibco, Invitrogen)
⦁ containing 10% FBS and 1% penicillin/streptomycin in a 37 °C humidified incubator with 5% CO2.
⦁ The medium was replaced every three days.
25
⦁ 2.2. MTT assay
⦁ The MTT assay was used to test retinoblastoma cell viability. Y79 and WERI-Rb-1 cells were
⦁ plated in 96-well plates at 5000 cells per well and were treated with different concentrations of
⦁ EPZ015666 (5 µM, 10 µM, 20 µM, 30 µM; Selleck Chemicals, Houston, TX, USA). After incubation
⦁ for different times (24 h, 48 h, and 72 h), 20 µl of MTT[3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2-H-
⦁ tetrazolium bromide] was added, and the cells were incubated for another 4 h. Cell viability was
⦁ represented by optical density (OD) values after dissolving cells in DMSO. Each experiment was
⦁ repeated three times.
34
⦁ 2.3. Colony formation assay
⦁ Y79 and WERI-Rb-1 cells were plated in six-well plates at 1000 cells per well and were treated
⦁ with various concentrations of EPZ015666, with the medium changed every 3 days. After culturing
⦁ for 14 days, the cells were stained for 1 h in 0.1% crystal violet dye. cell clusters with more than 50
⦁ cells were deemed as colonies and were counted. Each experiment was repeated three times.
⦁ 2.4. Cell cycle assay
⦁ Y79 and WERI-Rb-1 cells were treated with various concentrations of EPZ015666 for three days.
⦁ Prior to analysis, the cells were collected and were washed with PBS twice, fixed with 70% ethanol
⦁ at 4 °C for 1 h, and then again washed with PBS twice. The cell cycle was analysed via flow
⦁ cytometry (Becton Dickinson, San Jose, CA, USA) after incubation in a 500 µL mixture of PBS,

⦁ RNase A (20 μg/mL), and propidium iodide (PI, 50 μg/ml, Sigma-Aldrich, St. Louis, MO, USA) for
⦁ 20 min. The cells with 0 uM were deemed the control group. Each experiment was repeated three
⦁ times.
⦁ 2.5. Lentivirus acquisition and cell infection
⦁ The Coding sequence (CDS) of human PRMT5 gene was cloned into the PLVX-puro vector
⦁ (Addgene, Cambridge, MA,USA) and two different PRMT5 short hairpin RNAs (shRNAs) were
⦁ cloned into the pLKO.1-puro vector (Addgene). The lentivirus packaging vectors used were
⦁ packaging plasmid psPAX2 and envelope plasmid pMD2.G. The transfer vector PLVX-puro or
⦁ pLKO.1-puro together with the previous two plasmids were transfected into HEK-293T cells in a
⦁ suitable ratio mixed with the transfection reagent Lipofectamine 2000 (Thermo Fisher Scientific,
⦁ USA). The supernatant was filtered through a 0.45 μm filter (Millipore, Temecula, CA, USA) after
⦁ collection at 48 h, and the concentrated virus was then used to infect recipient cell lines with 8 mg/ml
⦁ polybrene. The stable lines were selected with 0.25 µg/ml puromycin for 2 weeks, and the cells were
⦁ collected and used to detect the protein level of PRMT5 via Western blot assay. The sequences
⦁ used are listed below: shPRMT5-1: 5’-GGGACTGGAATACGCTAATTG-3’; shPRMT5-2: 5’-
⦁ GCCATCACTCTTCCATGTTCT-3’.
17
⦁ 2.6. RNA extraction and real-time qPCR
⦁ RNA was extracted using TRIzol reagent (Invitrogen, Gergy Pontoise, France), reverse
⦁ transcription was conducted using a reverse transcriptase kit (TaKaRa, Kusatsu, Japan), and real-
⦁ time qPCR was detected via SYBR Premix Ex Taq (TaKaRa) according to the manufacturer’s
⦁ instructions. The qPCR reaction system consisted of 5 µl SYBR Premix Ex Taq, 1 µl forward primer,
⦁ 1 µl reverse primer, 2 µl cDNA, and 1 µl ddH2O. GAPDH was deemed the reference gene, and the
⦁ relative expression levels of target genes were calculated via the 2-ΔΔCT method. Each experiment
⦁ was repeated three times.

⦁ Table 1. The primer sequences for qPCR. Gene Sequence (5’–3’)
GAPDH Forward GCACCGTCAAGGCTGAGAAC

GAPDH PRMT5 PRMT5
27
Reverse Forward Reverse
TGGTGAAGACGCCAGTGGA TGAGGCCCAGTTTGAGATGCCTTA AGTAGCCGGCAAAGCCATGTAGTA

⦁ 2.7. Protein extraction and Western blot assay
⦁ Cells were collected and washed twice with PBS and lysed with a mixture of PARP buffer
⦁ (Beyotime Institute of Biotechnology, Beijing, China) and centrifuged at 12000 rpm for 20 min. The
⦁ protein concentration was determined via a BCA Protein Assay Kit (Beyotime, Shanghai, China).
⦁ Proteins were separated using 10% SDS-PAGE gels and transferred to 0.45 µm polyvinylidene
⦁ fluoride (PVDF) membranes (Millipore, Burlington, MA, USA). The PVDF membrane was incubated
⦁ in 5% nonfat milk at room temperature for 1 h and then incubated with PRMT5 (sc-376937, 1:500;
⦁ Santa Cruz Biotech, Dallas, TX, USA), P53 (sc-6243, 1:1000; Santa Cruz Biotech), P21 (12D1,
⦁ 1:1000; Cell Signaling Technology, Danvers, MA, USA), P27 (D69C12, 1:1000; Cell Signaling
⦁ Technology), CDK2 (78B2, 1:1000; Cell Signaling Technology), and β-actin antibody (20536-1-AP,
⦁ 1:5000; ProteinTech, Chicago, IL, USA) at 4 °C overnight. The membranes were washed three times
⦁ and then incubated with goat anti-rabbit (1:15000; ProteinTech) or goat anti-mouse (1:15000;
⦁ ProteinTech) secondary antibodies at room temperature for 1 h, the membranes were washed three

⦁ times. Then, the protein levels were detected using enhanced chemiluminescence (ECL). Each
⦁ experiment was repeated three times.
⦁ 2.8. Hematoxylin-eosin Staining and Immunohistochemistry
⦁ All human samples were collected in accordance with the Ethics Committee of Institute of the
⦁ Health Sciences. Retinoblastoma samples were obtained from surgery patients with retinoblastoma
⦁ at the eye hospital of Wenzhou medical university and the Third Affiliated Hospital of Nanchang
⦁ University, and colorectal cancer samples were obtained from surgery patients with colorectal
⦁ cancer at the Third Affiliated Hospital of Nanchang University. The samples were fixed with 10%
⦁ formalin and embedded in paraffin. The samples were sliced to sections before staining. The
⦁ prepared sections were stained with H&E to examine the structure or with immunohistochemistry for
⦁ the detection of PRMT5. For immunohistochemistry, the sections were baked at 60°C for 2 h. The
⦁ sample sections were deparaffinized with xylene and rehydrated with graded alcohol. The sections
⦁ were incubated with EDTA buffer (ZLI-9072, ZSGB-BIO, China) and boiled in a microwave oven for
⦁ 15 minutes. After incubation in goat serum (ZLI-9056, ZSGB-BIO, China), the sections were stained
⦁ with anti-PRMT5 (sc-376937, dilution 1:100, Santa Cruz Biotech, Dallas, TX, USA; ab-109451,
⦁ dilution 1:100, Abcam, Cambridge, UK) overnight at 4°C. The sections were washed four times with
⦁ PBS and then incubated with secondary antibody (KIT-5010, Maixin, China) at 37°C for 30 minutes.
⦁ The sections were washed four times with PBS. followed by staining with DAB (K5007, Dako,
⦁ Denmark) for visualization. Finally, the sections were counterstained with haematoxylin prior to
⦁ visualization. Staining of PRMT5 was scored by two independent pathologists. The score was
21 regarded as 0 (no), 1 (1-25%), 2 (26-50%), 3 (51-75%), 4 (76-100%) according to the percentage of
⦁ stained tissue area, and the intensity of staining was score on 0 (no), 1 (weak), 2 (medium), 3
⦁ (strong). The final score was calculated by multiplying the above two scores. In addition, the Clinical
⦁ and pathological characteristics of patients with RB were shown in Supplementary Table S1, the
⦁ table counted the patients’ gender, age, laterality, Invasion of Optic Nerve, Differentiation and
⦁ Choroidal Invasion.
⦁ 2.9. Xenotransplantation experiments
⦁ Animal experiments were processed in accordance with the ARVO Statement for the Use of
⦁ Animals in Ophthalmic and Vision Research. Twelve 4- to 5-week-old male athymic nude mice were
⦁ purchased from Shanghai Laboratory Animal Center (SLAC; Shanghai, China) for a subcutaneous
⦁ tumorigenesis test. Then, 0.2 ml PBS containing 1×107 Y79 cells was injected into the right axillary
⦁ subcutaneous region of the mouse. Two weeks after the subcutaneous cell injection, animals were
⦁ assessed for the successful transplantation of tumours. The mice were divided into two groups: a
⦁ vehicle group and an experimental group, with each group including six mice. The experimental
⦁ group received an oral dose of EPZ015666 at 150 mg/kg of body weight twice a day, and the vehicle
⦁ group was orally given an equal volume of vehicle. The body weights and the sizes of the tumours
⦁ were Measured every three days. At day 21 after treatment, the mice were sacrificed, and the
⦁ tumours were harvested for tumour size and weight measurement.
⦁ 2.10. Statistical analysis
⦁ All the data are presented as the means ± SD. The data were analyzed using Student’s t-test for
⦁ the in vitro study, and for the in vivo study, the differences in tumour volumes and body weights
⦁ between the experimental group and vehicle group were analysed by two-way ANOVA. A
⦁ P-value < 0.05 was regarded as statistically significant.

⦁ 3. RESULTS
⦁ 3.1. PRMT5 was overexpressed in retinoblastoma
⦁ PRMT5 is overexpressed in various cancers (SV et al., 2018; X et al., 2013; Y et al., 2017; JY et
⦁ al., 2018), so we analysed the expression of PRMT5 in retinoblastoma cells (Y79, WERI-Rb-1) and
⦁ tissues. The results show that the mRNA and protein levels of PRMT5 in retinoblastoma cells were
⦁ upregulated compared to retinal pigment epithelium cells (ARPE-19) (Figure 1A, 1B). To detect the
⦁ protein level of PRMT5 in human retinoblastoma tissues and adjacent tissues, two different
⦁ monoclonal antibodies were used. It has been reported that PRMT5 is positive in colorectal cancer(L
⦁ et al., 2017); therefore, for the positive control, we detected the expression of PRMT5 in colorectal
⦁ cancer. The results show that PRMT5 can be detected in colorectal cancer tissues, with PBS as a
⦁ negative control (Supplementary Figure 1), indicating that two monoclonal antibodies can effectively
⦁ detect PRMT5. Next, we detected PRMT5 expression human retinoblastoma tissues and adjacent
⦁ tissues. As shown in Figure 1C and 1D, the expression of PRMT5 was significantly increased in
⦁ retinoblastoma tissues. Western blotting also showed similar results (Figure 1E). Taken together,
⦁ these data imply that PRMT5 is overexpressed in retinoblastoma.
16
⦁ 3.2. Overexpressed PRMT5 promoted RB cell proliferation and knockdown of PRMT5
⦁ suppressed RB cell proliferation
⦁ The above results led us to investigate the function of PRMT5 in retinoblastoma. Based on
⦁ endogenous expression of PRMT5 in RB cell lines (Figure 1A, 1B), we constructed stable Y79 cells
⦁ with PRMT5 overexpression and stable WERI-Rb-1 cells with PRMT5 knockdown (Figure 2A, 2E).
⦁ The results of MTT and colony formation assays showed that the cell proliferation and colony
⦁ formation abilities were significantly increased after stably overexpression of PRMT5 (Figure 2B,
⦁ 2C). In addition, flow cytometry showed that the proportion of cells in G0/G1 phase decreased, and
⦁ the proportion of cells in S phase increased in stable Y79 overexpressing cells compared to control
⦁ cells (Figure 2D). While these effects of PRMT5 were reversed after treatment with EPZ015666.
⦁ Conversely, knockdown of PRMT5 in WERI-Bb-1 ARPE-19 and Y79 cells suppressed cell
⦁ proliferation (Figure 2F, Supplementary Figure 2), and knockdown of PRMT5 in WERI-Bb-1
⦁ suppressed colony formation ability (Figure 2G). Flow cytometry results showed that silencing
⦁ PRMT5 caused increase in G1 phase, decrease in S phase (Figure 2H).
31
⦁ 3.3. PRMT5 inhibitor EPZ015666 inhibits retinoblastoma cell proliferation
⦁ To demonstrate that EPZ015666 inhibits cell growth, we treated Y79 and WERI-Rb-1 cells with
⦁ different concentrations of EPZ015666 several time points (24 h, 48 h, 72 h) and assessed the cell
⦁ viability by MTT assay. EPZ015666 inhibited retinoblastoma cell proliferation in a concentration- and
⦁ time-dependent manner (Figure 3A, 3B). In addition, colony formation assays showed that the
⦁ number of colonies decreased as EPZ015666 dose increased (Figure 3C, 3D). To analyse the effect
⦁ of EPZ015666 on the cell cycle, we treated Y79 and WERI-Rb-1 cells with various concentrations of
⦁ EPZ015666 for three days and then analysed the change in the cell cycle via flow cytometry. The
⦁ results showed that the proportion of G1 phase cells increased and the proportion of S phase cells
⦁ decreased in retinoblastoma cells as the concentration of EPZ015666 increased (Fig. 3E, F).
⦁ 3.4. EPZ015666 regulates cell cycle-related factor expression
⦁ Changes in the cell cycle can affect cell proliferation, and cell cycle related factors play an
⦁ important role in the cell cycle. Based on the above results, EPZ015666 induced cell cycle arrest in
⦁ G1, and we investigated whether EPZ015666 affected the expression of key proteins regulating the
⦁ G1/S cell cycle transition. In our results, cell cycle-related factors were assessed by Western blot
⦁ assay. The results showed that the protein levels of CDK2 decreased, while the protein levels of
⦁ P21, P27, and P53, important anticancer factors, were increased after treatment of Y79 and WERI-

⦁ Rb-1 cells with EPZ015666 (Figure 4A, 4B). This indicates that the antiproliferative mechanism of
⦁ EPZ015666 is related to the protein levels of cell cycle-related factors in retinoblastoma cells.
⦁ 3.5. EPZ015666 inhibits tumour growth in vivo
⦁ To detect the role of EPZ015666 in vivo, cells were subcutaneously inoculated into BALB/c nude
⦁ mice, and tumour growth was monitored. The results showed that the vehicle group had more
⦁ proliferative ability than the experimental group. As shown in Figure 5A, 5B, the tumour sizes in the
⦁ vehicle group were larger than those in the experimental group. After 21 days of intervention, the
⦁ mice were sacrificed by euthanasia, and the xenografts were harvested. The tumour sizes and
⦁ weights of the experimental group were less than those of the vehicle group (Figure 5C, 5D). As
⦁ shown in Figure 5E, the weights of the mice were not significantly different between the experimental
⦁ group and vehicle group, indicating that no significant toxicity was found in the experimental group.
⦁ 4. Discussion
⦁ RB is the most common intraocular malignancy in children and it threatens patients’ vision and
⦁ lives to a great extent. Currently, chemotherapy is the first-line therapy in clinical practice (ID et al.,
⦁ 2018; A and PT, 2018). Nevertheless, various side effects are predictive of treatment failure. Thus,
⦁ new therapy strategies are urgently needed.
⦁ PRMT5 is a member of the PRMT family, and it plays an important role in cell signal transduction,
⦁ the cell cycle, and gene expression by regulating histone and non-histone methylation (W et al.,
⦁ 2019; N et al., 2015). An increasing number of studies have found PRMT5 overexpression in various
⦁ cancers; for example, overexpression of PRMT5 in hepatocellular carcinoma promotes
⦁ hepatocellular carcinoma growth by regulating β-catenin, which indicates that PRMT5 plays the role
⦁ of an oncogene in hepatocellular carcinoma (B et al., 2015). PRMT5 promotes human lung cancer
⦁ cell growth by targeting the PRMT5/Akt signalling axis (S et al., 2019), and the overexpression of
⦁ PRMT5 is related to tumour phenotype and poor prognosis in ovarian cancer; depleting PRMT5 by
⦁ siRNA inhibits the growth and proliferation of ovarian cancer cells and induces apoptosis in vitro (X
⦁ et al., 2013). Similarly, we found that PRMT5 is overexpressed in retinoblastoma cell lines (Y79,
⦁ WERI-Rb-1) compared to retinal pigment epithelium cells (ARPE-19) at the mRNA and protein
⦁ levels. In addition, although the sensitivity of the two antibodies is different, our results show that
⦁ PRMT5 is overexpressed in retinoblastoma tissues compared to adjacent tissues. Additionally,
⦁ overexpression of PRMT5 promotes Y79 proliferation and colony formation. Conversely, knockdown
⦁ of PRMT5 inhibits WERI-Rb-1 proliferation and colony formation, and knockdown of PRMT5 inhibits
⦁ cell proliferation in ARPE-19 and Y79 cells. Moreover, PRMT5 regulates the retinoblastoma cell
⦁ cycle.
⦁ EPZ015666 is a common PRMT5 inhibitor, and its antitumour effects have been confirmed in
⦁ various cancers. Similarly, our research indicates that EPZ015666 inhibits Y79 and WERI-Rb-1
⦁ proliferation in a dose- and time-dependent manner. Consistent with this, EPZ015666 also
⦁ significantly inhibited retinoblastoma growth in vivo.
⦁ The cell cycle includes G0/G1, S, G2, and M phases, where the G0/G1 phase is a protein
⦁ synthesis stage that prepares for chromosome synthesis, which occurs in the S phase, and
⦁ represents cell proliferation. In the G2 phase, protein synthesis occurs to prepare for cell division,
⦁ and in the M phase, cell division occurs. In our studies, after Y79 and WERI-Rb-1 cells were treated
⦁ with EPZ015666, cell cycle analysis showed that the proportion of cells in G1 phase increased, while
⦁ the proportion of cells in S phase decreased, as determined by flow cytometry. This indicates that
⦁ EPZ015666 induced Y79 and WERI-Rb-1 cell cycle G0/G1 arrest, thereby inhibiting cell proliferation.

⦁ Cyclin-dependent kinase (CDK) is an important factor in cell cycle regulation. CDK2 is a member
⦁ of the CDK family that activates the G1 to S transition. Cyclin-dependent kinase inhibitors (CDKIs)
⦁ are another important factor in the cell cycle, regulating the cell cycle via CDK (K et al., 2019a; Z et
⦁ al., 2016; S et al., 2017). In our research, we found that the expression of the tumor suppressor
⦁ genes p53, p21, and p27 was upregulated, while CDK2 was downregulated after treatment with
⦁ EPZ015666 in RB cells. This result was supported by previous studies that p27, from the p21 family,
⦁ is an inhibitor of CDK, and the increased expression inhibited the activity of CDK (H and T, 1994).
⦁ The tumour suppressor gene p53 is related to various cell biological functions, including signal
⦁ transduction, posttranscriptional regulation, the cell cycle, and cell apoptosis. Many studies have
⦁ found p53 mutations in various cancers, including breast cancer (N et al., 2019), gastric cancer, and
⦁ hepatocellular carcinoma (G and M, 2019). In RB, p53 is normal while its activity is inhibited, but
⦁ once p53 is activated, p21 and p27 are upregulated and induce cell cycle arrest by inhibiting CDK
13 (JL et al., 2017).
⦁ In summary, our studies show that PRMT5 is overexpressed in RB and promotes RB cell growth.
⦁ The PRMT5 inhibitor EPZ015666 inhibits RB growth in vitro and in vivo. Thus, our studies provide a
⦁ potential strategy for the treatment of RB.
⦁ Acknowledgements
⦁ This research was mainly supported by the Jiangxi Provincial Natural Science Foundation of
⦁ China (No. 20202ACB206005 to GFH and No. S2016QNJJB0718 to GFH), and the National Natural
⦁ Science Foundation of China (No. 81560158 to GFH).

⦁ Authors’ Contributions: XL and JZH finished most of the experiments in this study and wrote the
⦁ manuscript, XL and JZH have the same contribution in this study. LBM, YY, and WWC finished the
⦁ animal experiments, SJD, ALJ, and YG arranged and analysed the data, YS provided the technical
⦁ support, and GFH designed, managed and supported all of the study.
25
⦁ Competing Interests
⦁ None of the authors have competing interests to declare.
28
29 References
30
31 A, M. & PT, F. (2018). Cancers of the eye. Cancer metastasis reviews 37(4): 677-690.
32 AB, D., MT, F., AH, N., JM, P., I, B., CA, S., PR, S., YK, T., KL, B., LE, H., SC, C., L, D., DL, F. & A, R.
⦁ (2019). Rabbit Model of Intra-Arterial Chemotherapy Toxicity Demonstrates Retinopathy and
⦁ Vasculopathy Related to Drug and Dose, Not Procedure or Approach. Investigative ophthalmology &
35 visual science 60(4): 954-964.
⦁ AS, T., OO, P., MV, Z., AR, M., TE, K., OS, E., AI, S., MV, Z., YV, T., KV, L., VS, S., GB, S. & BV, A.
⦁ (2019). Safe and Effective Delivery of Antitumor Drug Using Mesenchymal Stem Cells Impregnated
⦁ with Submicron Carriers. ACS applied materials & interfaces 11(14): 13091-13104.
⦁ B, H., A, B.-S., JT, S., KA, L., AJ, S., RN, N. & J, G. (2019). The protein arginine methyltransferase
⦁ PRMT5 confers therapeutic resistance to mTOR inhibition in glioblastoma. Journal of neuro-
41 oncology 145(1): 11-22.
⦁ B, Z., S, D., Z, L., L, L., S, Z., X, C., X, C. & Y, W. (2015). Targeting protein arginine
⦁ methyltransferase 5 inhibits human hepatocellular carcinoma growth via the downregulation of beta-
⦁ catenin. Journal of translational medicine 13(undefined): 349.
⦁ B, Z., S, Z., L, Z., X, C., Y, Z., L, C., J, Z., X, W., X, Z. & N, M. (2017). Arginine methyltransferase

⦁ inhibitor 1 inhibits gastric cancer by downregulating eIF4E and targeting PRMT5. Toxicology and
⦁ applied pharmacology 336(undefined): 1-7.
⦁ CL, S., EM, F., JD, A., C, A., M, P., P, R., S, K., CG, B. & JA, S. (2013). Retinoblastoma frontiers with
⦁ intravenous, intra-arterial, periocular, and intravitreal chemotherapy. Eye (London, England) 27(2):
5 253-264.
⦁ D, C., S, Z., M, H., H, X., L, L. & X, Y. (2017). Role of protein arginine methyltransferase 5 in
⦁ inflammation and migration of fibroblast-like synoviocytes in rheumatoid arthritis. Journal of cellular
⦁ and molecular medicine 21(4): 781-790.
⦁ D, S., U, H., M, S., K, J., S, S. & S, A. (2016). Toxic effects of melphalan, topotecan and carboplatin
⦁ on retinal pigment epithelial cells. Acta ophthalmologica 94(5): 471-478.
11 E, C.-P., KG, K., CR, M., PA, B.-S., TJ, W., LD, J., N, R., MJ, M., L, J., SL, J., KA, W., T, L., K, S., SA,
12 R., A, R., MP, S., NJ, W., RM, P., JJ, S., O, B., M, P., TF, H., K, N., KP, O., KT, G., R, K., MP, M., RA,
⦁ C., R, C. & KW, D. (2015). A selective inhibitor of PRMT5 with in vivo and in vitro potency in MCL
⦁ models. Nature chemical biology 11(6): 432-437.
⦁ G, D. O. & M, C. (2019). Mutant p53 and Cellular Stress Pathways: A Criminal Alliance That
⦁ Promotes Cancer Progression. Cancers 11(5): undefined.
⦁ G, H., X, W., Y, H. & P, W. (2018). Protein arginine methyltransferase 5 promotes bladder cancer
⦁ growth through inhibiting NF-kB dependent apoptosis. EXCLI journal 17(undefined): 1157-1166.
19 H, L., L, K., J, J., A, C., C, P., O, T., L, C., M, D.-A., N, H., I, T. & M, L. R. (2019). LKB1 regulates
⦁ PRMT5 activity in breast cancer. International journal of cancer 144(3): 595-606.
⦁ H, T. & T, H. (1994). p27, a novel inhibitor of G1 cyclin-Cdk protein kinase activity, is related to p21.
22 Cell 78(1): 67-74.
⦁ ID, F., Z, O., E, K., C, D., T, C., I, S., SI, O., MA, R. & MS, S. (2018). The management of
⦁ retinoblastoma. Oncogene 37(12): 1551-1560.
⦁ J, C., V, K., S, T., F, Y., P, S., LW, A., C, A., S, P., GV, D., RA, B. & S, S. (2013). Protein arginine
⦁ methyltransferase 5 (PRMT5) inhibition induces lymphoma cell death through reactivation of the
⦁ retinoblastoma tumor suppressor pathway and polycomb repressor complex 2 (PRC2) silencing. The
⦁ Journal of biological chemistry 288(49): 35534-35547.
⦁ JL, Q., WY, S., ZF, C., LF, Z., QP, Q., YC, Y. & H, L. (2017). Oxoaporphine Metal Complexes (Co, Ni,
⦁ Zn) with High Antitumor Activity by Inducing Mitochondria-Mediated Apoptosis and S-phase Arrest in
⦁ HepG2. Scientific reports 7(undefined): 46056.
⦁ JY, J., JS, L., ER, P., YN, S., MY, K., HJ, S., HY, J., EH, C., SM, M., US, S., SH, P., CJ, H., DW, C.,
⦁ MB, G., SB, K. & KH, L. (2018). Protein arginine methyltransferase 5 is implicated in the
⦁ aggressiveness of human hepatocellular carcinoma and controls the invasive activity of cancer cells.
⦁ Oncology reports 40(1): 536-544.
⦁ K, G., H, L., Z, Z., F, W., P, H., X, P., J, F., H, C., G, S. & P, O. (2019a). The Molecular Mechanisms
⦁ of Protective Role of Se on the G/G Phase Arrest Caused by AFB in Broiler's Thymocytes. Biological
⦁ trace element research 189(2): 556-566.
⦁ K, Z., J, S., H, T., X, Y., C, L., H, Z. & W, D. (2019b). Rational Design, synthesis and biological
⦁ evaluation of novel triazole derivatives as potent and selective PRMT5 inhibitors with antitumor
⦁ activity. Journal of computer-aided molecular design 33(8): 775-785.
42 L, P., H, W., L, C., Ö, D., G, S., E, S., J, W., J, M., L, Z., M, F., A, S., R, A., M, K., ZY, Z. & T, L.
⦁ (2017). Adapting AlphaLISA high throughput screen to discover a novel small-molecule inhibitor
⦁ targeting protein arginine methyltransferase 5 in pancreatic and colorectal cancers. Oncotarget
45 8(25): 39963-39977.
⦁ MA, C., TD, N. & JH, B. (2019a). Estimation of Kidney Function in Oncology: Implications for
⦁ Anticancer Drug Selection and Dosing. Clinical journal of the American Society of Nephrology :
48 CJASN 14(4): 587-595.
49 MA, L., JP, C., Á, R., C, S., L, C., JC, R., R, C., S, M.-M., C, C., B, C., E, G.-M., JAG, A., C, A., K, P.,
⦁ DD, C., NM, V. & LA, Q. (2019b). Association Study Among Candidate Genetic Polymorphisms and
⦁ Chemotherapy-Related Severe Toxicity in Testicular Cancer Patients. Frontiers in pharmacology
⦁ 10(undefined): 206.
⦁ N, I., NI, K., M, A. & EH, C. (2019). Transfection with p21 and p53 tumor suppressor plasmids

⦁ suppressed breast tumor growth in syngeneic mouse model. Gene 701(undefined): 32-40.
⦁ N, S., JE, K. & D, S. (2015). The PRMT5 arginine methyltransferase: many roles in development,
⦁ cancer and beyond. Cellular and molecular life sciences : CMLS 72(11): 2041-2059.
⦁ P, J., N, Z., M, Y., Y, Z., Z, Z., J, S., Z, W., J, Z. & Z, G. (2018). Protein arginine methyltransferase 5
⦁ promotes lung cancer metastasis via the epigenetic regulation of miR-99 family/FGFR3 signaling.
⦁ Cancer letters 427(undefined): 38-48.
7 S, K., F, L., KJ, V., G, G., P, D., LF, N., K, L., Y, Z., Y, L., V, G., MT, B., SD, N. & MA, S. (2018).
⦁ Genetic deletion or small-molecule inhibition of the arginine methyltransferase PRMT5 exhibit anti-
⦁ tumoral activity in mouse models of MLL-rearranged AML. Leukemia 32(2): 499-509.
⦁ S, M., FS, S., N, B. & Y, Y. (2017). Indole-3-carbinol induces G1 cell cycle arrest and apoptosis
⦁ through aryl hydrocarbon receptor in THP-1 monocytic cell line. Journal of receptor and signal
⦁ transduction research 37(5): 506-514.
⦁ S, P., SN, V., H, E.-B., P, T. & S, S. (2004). Human SWI/SNF-associated PRMT5 methylates histone
⦁ H3 arginine 8 and negatively regulates expression of ST7 and NM23 tumor suppressor genes.
⦁ Molecular and cellular biology 24(21): 9630-9645.
⦁ S, Z., Y, M., X, H., Y, Z. & X, C. (2019). Targeting PRMT5/Akt signalling axis prevents human lung
⦁ cancer cell growth. Journal of cellular and molecular medicine 23(2): 1333-1342.
18 SV, G., WA, K., C, T., MB, P., XP, Z., R, M. d. O., E, P., K, D., A, B.-S., B, L., C, M., MT, M., C, C., N,
⦁ J., RG, K. & O, B. (2018). Activation of the p53-MDM4 regulatory axis defines the anti-tumour
⦁ response to PRMT5 inhibition through its role in regulating cellular splicing. Scientific reports 8(1):
21 9711.
⦁ TN, C., AA, D., E, A., AC, Z., MD, S. & DR, H. (2012). The immunoreceptor tyrosine-based activation
⦁ motif (ITAM) -related factors are increased in synovial tissue and vasculature of rheumatoid arthritic
⦁ joints. Arthritis research & therapy 14(6): R245.
⦁ W, X., X, C., L, L., Y, S., M, Z. & Y, Z. (2019). Role of protein arginine methyltransferase 5 in human
⦁ cancers. Biomedicine & pharmacotherapy = Biomedecine & pharmacotherapie 114(undefined):
27 108790.
⦁ X, B., S, Z., T, L., Y, L., Y, L. & X, Y. (2013). Overexpression of PRMT5 promotes tumor cell growth
⦁ and is associated with poor disease prognosis in epithelial ovarian cancer. The journal of
⦁ histochemistry and cytochemistry : official journal of the Histochemistry Society 61(3): 206-217.
⦁ X, L., C, W., H, J. & C, L. (2019). A patent review of arginine methyltransferase inhibitors (2010-
⦁ 2018). Expert opinion on therapeutic patents 29(2): 97-114.
⦁ Y, A., D, M., T, Y., T, T., H, N., Y, M. & T, N. (2018). Expression of protein arginine methyltransferase-
⦁ 5 in oral squamous cell carcinoma and its significance in epithelial-to-mesenchymal transition.
⦁ Pathology international 68(6): 359-366.
⦁ Y, W., Z, W., J, Z. & R, L. (2017). Elevated expression of protein arginine methyltransferase 5
⦁ predicts the poor prognosis of breast cancer. Tumour biology : the journal of the International Society
⦁ for Oncodevelopmental Biology and Medicine 39(4): 1010428317695917.
⦁ Z, F., Y, G., M, Q., X, Q., Y, D., S, W., Z, Q., Y, L. & J, O. (2016). Down-regulation of dihydrofolate
⦁ reductase inhibits the growth of endothelial EA.hy926 cell through induction of G1 cell cycle arrest
⦁ via up-regulating p53 and p21(waf1/cip1) expression. Journal of clinical biochemistry and nutrition
42 58(2): 105-113.
⦁ Z, L., H, J., L, H., Y, Z., W, S., X, R., W, Z., L, L., M, L., J, Z. & J, C. (2018a). Prognostic impact of
⦁ SET domain-containing protein 8 and protein arginine methyltransferase 5 in patients with
⦁ hepatocellular carcinoma following curative resection. Oncology letters 16(3): 3665-3673.
46 Z, L., J, Z., X, L., S, L., Q, W., Chen, D., Z, H., T, Y., J, D., J, L., M, Y., J, F., S, H., Q, G., Y, Z. & X, H.
⦁ (2018b). The LINC01138 drives malignancies via activating arginine methyltransferase 5 in
⦁ hepatocellular carcinoma. Nature communications 9(1): 1572.
⦁ Z, W., J, K., Y, W., J, Z., T, W., N, L., J, F., H, W., J, Z. & R, L. (2018c). PRMT5 determines the
⦁ sensitivity to chemotherapeutics by governing stemness in breast cancer. Breast cancer research
51 and treatment 168(2): 531-542.
52

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Highlights
⦁ There is an urgent need for novel strategies or drugs for retinoblastoma.
⦁ Protein arginine methyltransferase 5 plays an important role in various cancers.
⦁ Protein arginine methyltransferase 5 promotes retinoblastoma cell proliferation.
⦁ EPZ015666, a selective PRMT5 inhibitor, exerts antitumour effects in retinoblastoma.GSK3235025