PRMT5 silencing selectively affects MTAP‐deleted mesothelioma: In vitro evidence of a novel promising approach

Abstract Malignant mesothelioma (MM) is an aggressive asbestos‐related cancer of the serous membranes. Despite intensive treatment regimens, MM is still a fatal disease, mainly due to the intrinsic resistance to current therapies and the lack of predictive markers and new valuable molecular targets. Protein arginine methyltransferase 5 (PRMT5) inhibition has recently emerged as a potential therapy against methylthioadenosine phosphorylase (MTAP)‐deficient cancers, in which the accumulation of the substrate 5'‐methylthioadenosine (MTA) inhibits PRMT5 activity, thus sensitizing the cells to further PRMT5 inhibition. Considering that the MTAP gene is frequently codeleted with the adjacent cyclin‐dependent kinase inhibitor 2A (CDKN2A) locus in MM, we assessed whether PRMT5 could represent a therapeutic target also for this cancer type. We evaluated PRMT5 expression, the MTAP status and MTA content in normal mesothelial and MM cell lines. We found that both administration of exogenous MTA and stable PRMT5 knock‐down, by short hairpin RNAs (shRNAs), selectively reduced the growth of MTAP‐deleted MM cells. We also observed that PRMT5 knock‐down in MTAP‐deficient MM cells reduced the expression of E2F1 target genes involved in cell cycle progression and of factors implicated in epithelial‐to‐mesenchymal transition. Therefore, PRMT5 targeting could represent a promising new therapeutic strategy against MTAP‐deleted MMs.


| INTRODUC TI ON
Malignant mesothelioma (MM) is a very aggressive tumour of the serous membranes, with the most frequent type developing in the pleura, the membrane covering the lungs and lining the chest cavity. MM is mainly associated with asbestos exposure and is the most common type of asbestos-related cancer. 1,2 Despite current intensive treatment regimens, including chemotherapy, radiation and surgery, MM patients' overall survival (OS) range from 8 to 36 months, depending on the stage and the histological subtype. 3,4 Many efforts have been made in the past years to obtain better results in terms of survival, but with limited results. [5][6][7][8] Only recently, in a phase 3 trial (NCT00651456), the addition of bevacizumab to pemetrexed plus cisplatin achieved an improvement of OS. 9 Therefore, a better understanding of the molecular machinery underlying MM development is required to move forward against this malignancy. In particular, the discovery of crucial pathways for MM cell proliferation and spreading and the identification of new druggable targets and predictive markers for the stratification of patients who could benefit from a specific therapy are fundamental goals of pre-clinical research aimed to the improvement of patients' outcome.
Among the most common genetic alterations or deregulated pathways identified in MM, deletions in the cyclin-dependent kinase inhibitor 2A (CDKN2A) locus, inactivation of the retinoblastoma (RB) pathway, mutations in the BRCA1-associated protein 1 (BAP1) and neurofibromatosis type 2 (NF2) genes, and aberrant regulation of phosphatidylinositol-4,5-bisphosphate 3-kinase (PI3K)/AKT pathway are all related to MM uncontrolled growth and resistance to treatment-induced cell death. [10][11][12] Moreover, MM poor prognosis has also been found to be related to the activation of the epithelial-to-mesenchymal transition (EMT) programme 13,14 a process involving genetic, epigenetic and morphological changes in epithelial cells, leading to acquisition of a fibroblast-like cell morphology, reduction in cell adhesion and gain of cell motility, which promotes migration and invasion. 15,16 Deletion on the short arm of chromosome 9, including the CDKN2A locus, is one of the first and most common mutations described in MM. 17 The discovery that CDKN2A deletion in cancer cells commonly involves codeletion of adjacent genes opened new perspectives in cancer research with a possible impact also for MM 18 It has indeed observed that the methylthioadenosine phosphorylase (MTAP) gene, encoding a key enzyme in the adenosine and methionine salvage pathway from the substrate 5'-methylthioadenosine (MTA), is frequently codeleted with CDKN2A in different cancer types 19 including MM 20,21 The MTAP gene has been suggested to be a tumour suppressor, the loss of which results in a higher cell invasive potential and poor prognosis for patients with different cancer types. 22 Importantly, MTAP loss determines the accumulation of the MTA substrate, a natural inhibitor of protein arginine methyltransferase 5 (PRMT5), thus generating a hypomorphic PRMT5 state in MTAP-deficient cancers that are, in this way, selectively sensitized to further PRMT5 inhibition. This vulnerability can be exploited therapeutically, and PRMT5 targeting in MTAP-deficient cancers has indeed become the focus of recent research. [23][24][25] PRMT5 belongs to a family of ten protein arginine methyltransferases (PRMTs) ubiquitously expressed in mammalian cells, which methylate arginine residues on histones and other proteins, although their biological role is still underexplored. PRMT5 regulates a broad range of physiological and cancer-associated processes, such as DNA damage response, apoptosis control, EMT and inflammation, and is involved in the inhibition of tumour suppressors, including RB proteins, p53, programmed cell death 4 (PDCD4) and activation of survival pathways such as PI3K/AKT axis26-29 Overall, these considerations prompted us to investigate whether PRMT5 could be a valuable MM therapeutic target, the inhibition of which could impact on pathways fundamental for MM biology.

| Immunohistochemical analysis
Formalin-fixed, paraffin-embedded tumour specimens were used for tissue microarray (TMA) construction. Multi-tissue pleural mesothelioma arrays were obtained from the Section of Pathology, Siena Hospital, Siena, Italy, and the Anatomy and Pathology Unit, Ospedale dei Colli, AORN, 'Monaldi', Naples, Italy, and consisted of 2-mm representative areas of resected tumour and normal pleura controls. From each tissue microarray, 4-μm-thick paraffin sections were prepared for immunohistochemistry. Clinical information about mesothelioma specimens is summarized in Table S1.
Based on the expression patterns identified in the resection specimens, the tumour cell staining in TMA was evaluated in comparison with normal pleura. Two pathologists blinded to the clinical data evaluated the staining of each specimen. To avoid inter-observer variability, the mean value of the scores was adapted for further analysis. The primary rabbit polyclonal anti-PRMT5 antibody (Abcam, Cambridge, UK, Cat #ab109451, RRID:AB_10863428) at 1:70 dilutions was used according to the manufacturer's instructions.
The assessment of PRMT5 expression levels included the staining intensity and the percentage of stained cells. PRMT5 was analysed for both nuclear and cytoplasmic staining. The staining intensity was scored as 0 = no staining, 1 = moderate expression and 2 = strong expression; the results were categorized according to the following distribution: 0 =< 10%, 1 = 10% -50% and 2 ≥ 50% staining. The PRMT5 expression score was determined as a combined score of staining intensity and distribution. Samples with a final immunoscore ≥ 2 were considered as PRMT5-positive.  (Table S2). All cell lines were routinely passaged every 1-2 weeks.

| Cell lines and culture conditions
Mesothelial origin of primary cell cultures was assessed by haematoxylin/eosin, calretinin and WT-1 staining.

| Fluorescence in situ hybridization (FISH)
Qualitative detection of CDKN2A gene (green signal) deletions and the classical satellite III region of chromosome 9 (CEN9) (red signal) were detected by fluorescence in situ hybridization (FISH), using SPEC CDKN2A/CEN 9 Dual Color Probe ZytoLight (ZytoVision GmbH, Bremerhaven, Germany). Cell lines were fixed with 4% paraformaldehyde for 10 minutes at RT, washed with PBS and pre-treated with heat pre-treatment solution citric, following the manufacturer's protocol. FISH was performed with a hybridization automation (HYBrite; Abbott Molecular). Probe was placed on the samples, covered with a glass slide and then sealed with rubber cement. After codenaturation at 78°C for 10 minutes, the probe and the target DNA were allowed to hybridize at 37˚C overnight in a humid and dark atmosphere. Next day, the excess of probe was washed in 25× wash buffer A at 37°C. Slides were air-dried in the darkness and counterstained with 4',6-diamidino-2-phenylindole (DAPI). Analysis was performed using a fluorescent microscope (Leica, Wetzlar, Germany) and Leica LAS v3.8 Software (Leica) at ×630 magnification, equipped with SpectrumGreen™, SpectrumOrange™ filters. For each specimen, at last 100 intact non-overlapping nuclei with good signals were required for valid scoring.

| LC-MS/MS measurement of MTA intracellular content
The protocol used is based on the method published by Stevens and was used. LC separation was carried out using a mobile phase consisting of 0.1% acetic acid in water (Solvent A) and 0.1% acetic acid in acetonitrile (Solvent B). The gradient employed was as follows: 0-2 minutes isocratic 5% solvent B, 2-10 minutes linear increase from 5% to 100% solvent B, hold at 100% solvent B for 3 minutes, and 3 minutes post-run equilibration. The flow rate was set to 500 µL/ min. Sample volumes of 20 µL were injected. The API 4000 mass spectrometer was operated in positive mode using turbo ion spray with the following parameters: gas 1 as 55, gas 2 as 45 and the curtain gas as 17 (all arbitrary units).
The turbo ion spray source was heated to 450°C. The declustering potential was set to 47 and the entrance potential to 9.4 V.

| Protein extraction and Western Blotting
For Western blotting analysis, cells were harvested on ice and lysed as previously described 16 Equal amounts of proteins

| Statistical analysis
Statistical analyses were performed using the GraphPad Prism

| PRMT5 expression in MM and normal mesothelial cells
To explore the role of PRMT5 in MM, we first analysed its expression   Table 1.

| PRMT5 knock-down decreases proliferation of MTAP-negative MM cells
To

| PRMT5 knock-down impacts on the expression of epithelial-to-mesenchymal transition markers in MTAP-negative MM cells
EMT is a cellular process whereby epithelial cells lose some of their typical characteristics and acquire mesenchymal properties. 40 This phenomenon has a key role in cancer motility and drug resistance and has recently been demonstrated to be promoted by PRMT5. 41 Therefore, we investigated the effect of PRMT5 knock-down in the MTAP-negative cells, NCI-H2452, on the expression of EMT markers. As shown in Figure 6, PRMT5 depletion led to the up-regulation  To analyse at the molecular level the growth-promoting activity of PRMT5 in MM cells, we first focused on the E2F pathway. We observed that PRMT5 knock-down led to a decreased expression of E2F1 target genes in MTAP-negative cells. E2F1 has previously been found to exert controversial roles in different human cancers, promoting either proliferation or tumour suppression, depending on the context 50 Although E2F1 has been proposed as a therapeutic target and it is an independent prognostic factor for many cancers, [51][52][53][54] the role of E2F1 in MM has poorly been investigated. Among other mechanisms, E2F1 can be regulated by PRMT5 arginine methylation;

| D ISCUSS I ON
it has indeed been demonstrated that the binding of cyclin A to E2F1 augments PRMT5 methylation of E2F1, thus ensuring that it is locked in a cell cycle progression mode 55 Based on the observed down-regulation of cyclin A, together with the down-regulation of other E2F1 target genes, in PRMT5-silenced MM cells, it could be hypothesized that the E2F1 activity in MM is affected by PRMT5-mediated arginine methylation. However, this needs further investigation.
Interestingly, among the E2F1 target genes down-regulated by PRMT5 silencing and involved in cell cycle progression, the EZH2 gene encodes a factor implicated in the epigenetic regulation of gene expression 56 which is a negative prognostic factor and a potential therapeutic target in MM 57 Finally, we identified a subset of EMT markers regulated by PRMT5 in MM, including E-cadherin, N-cadherin, α-smooth muscle actin and MMP9, indicating that targeting PRMT5 could hamper this crucial process for cancer progression.

| D ISCUSS I ON
To the best of our knowledge, this is the first study extensively exploring the possible therapeutic potential of targeting PRMT5 in MTAP-deleted MM cells, also showing the effect of its silencing on the expression of genes implicated in MM growth and progression.
In agreement with previous studies underlining the potential value of PRMT5 as a therapeutic approach, 58-61 our results represent a starting point for the evaluation of PRMT5 inhibitors also in MTAPdeleted mesothelioma.

This work was supported by Mesothelioma Applied Research
Foundation (GRANT ID 483418).

CO N FLI C T S O F I NTE R E S T
The authors have no conflicts of interest to declare.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that supports the findings of this study are available in the supplementary material of this article.