ANTISENSE OLIGONUCLEOTIDES AND THIER USE FOR THE TREATMENT OF CANCER

Abstract

The present invention concerns the treatment of prostate cancer and particularly castration resistant prostate cancer (CRPC). The Heat Shock Protein Hsp27, a chaperone protein, has been long demonstrated as a driver of Castration Resistance Prostate Cancer (CRPC). In the light of identification of the molecular mechanisms, the inventor determined that the Probable ATP-dependent RNA helicase DDX5 is an interactor of Hsp27 and DDX5's expression is modulated by Hsp27. They confirmed that DDX5 overexpression is correlated to the aggressiveness of the tumor, to the CRPC emergency and to the biochemical recurrence risk. They also developed DDX5-targeting antisense oligonucleotides for research purpose and clinical application. Thus, the invention relates to an inhibitor of DDX5 wherein said inhibitor reduces the expression and/or activity of DDX5 in a subject in need thereof and targets the gene or the mRNA of DDX5.

Claims

1. A method of treating cancer in a subject in need thereof, comprising administering to the subject a therapeutically effective amount of an inhibitor of DDX5, wherein said inhibitor reduces expression and/or activity of DDX5 and targets the gene or mRNA of DDX5.

2. The method according to claim 1 wherein said inhibitor reduces the expression and/or the activity of DDX5 and targets the nucleic acids sequence SEQ ID NO: 94.

3. The method according to claim 1 wherein said inhibitor reduces the expression and/or the activity of DDX5 and targets at least from 15 to 25 nucleic acids of SEQ ID NO: 94.

4. The method according to claim 1 wherein said inhibitor reduces the expression and/or the activity of DDX5 and targets at least a region comprising nucleic acids 276-515, or 1056-1155 or 1396-1795 or 1856-1955 of SEQ ID NO: 94.

5. The method according to claim 1 wherein said inhibitor is a siRNA, a shRNA, an antisense oligonucleotide, miRNA or a ribozyme.

6. The method according to claim 3 wherein said inhibitor is an antisense oligonucleotide selected from the group consisting of: SEQ ID NO:1 to SEQ ID NO:93.

7. The method according to claim 5 wherein said inhibitor is an antisense oligonucleotide set forth as SEQ ID NO:3.

8. A vector comprising a heterologous nucleic acid, wherein the heterologous nucleic acid encodes an inhibitor according to claim 1.

9. (canceled)

10. The method according to claim 1, wherein the cancer is selected from the group consisting of adrenal cortical cancer, anal cancer, bile duct cancer, bladder cancer, bone cancer, brain and central nervous system cancer, breast cancer, Castleman disease, cervical cancer, colorectal cancer, endometrial cancer, esophagus cancer, gallbladder cancer, gastrointestinal carcinoid tumors, Hodgkin's disease, non-Hodgkin's lymphoma, Kaposi's sarcoma, kidney cancer, laryngeal and hypopharyngeal cancer, liver cancer, lung cancer, mesothelioma, plasmacytoma, nasal cavity and paranasal sinus cancer, nasopharyngeal cancer, neuroblastoma, oral cavity and oropharyngeal cancer, ovarian cancer, pancreatic cancer, penile cancer, pituitary cancer, prostate cancer, retinoblastoma, rhabdomyosarcoma, salivary gland cancer, skin cancer, stomach cancer, testicular cancer, thymus cancer, thyroid cancer, vaginal cancer, vulvar cancer, and uterine cancer.

11. The method according to claim 1, wherein the cancer is a prostate cancer, a resistant prostate cancer or a Castration Resistant Prostate Cancer (CRPC).

12. A pharmaceutical composition which comprises an inhibitor according to claim 1.

13. (canceled)

14. The method according to claim 9, wherein the cancer is a chemotherapy or radiotherapy resistant prostate cancer.

Description

FIGURES

[0185] FIG. 1. Developing Anti Sense Oligonucleotide (ASO) inhibiting DDX5 protein levels in dose-dependency manner

[0186] A, The first screening to define the ASO inhibiting DDX5 protein expression. The hASO #51 was found to decrease DDX5 protein level in dose-dependency manner. B, The hASO #51 inhibited cell proliferation of the DU 145 cells. The MTT test was done with biological triplication. ** The data was analyzed using independent sample t-test.t value=5.513, p=0.00528. C and D, The second screening testing 8 different bi-specific ASO (hmASOs) along with the hASO #51 as a positive control. ASO #3, #55, #71 displayed to deplete DDX5 protein levels significantly. Further experiment showing that hmASO #3 is able to decrease DDX5 expression.

[0187] FIG. 2: Testing different batches of ASO3 and ASO51 to inhibit DDX5 protein levels.

[0188] ASO3 old/new/vivo are the same ASO3 (SEQ ID NO:3) and from just different synthesis batches. ASO51 old/new are the same ASO51 (SEQ ID NO:51) and from just different synthesis batches. The protocol for ASO transfection is the same as reported in Material & Methods.

[0189] FIG. 3: WB testing various ASOs that are nearby the ASO3.

[0190] ASO3 (SEQ ID NO:3) and nearby ASO3 are tested at a concentration at 120 nM. ASO3-1, ASO+1 and ASO3+3 also inhibit DDX5 protein level beside the ASO3.

EXAMPLE

[0191] Material & Methods

[0192] Cell Lines and Cell Culture Condition

[0193] We had different human prostate cell lines in our study: the normal phenotype cell PNT1A (ECACC, European Collection of Cell Cultures, England), castration sensitive (CS) LNCaP (ATCC, American Type Culture Collection (Rockville, Md., USA)), and castration-resistant (CR) DU145 and PC3 cell lines (ATCC). We also used two LNCaP-derived cell lines, LNCaP Mock and LNCaP Hsp27 which were established as described (Baylot et al. 2012). The cells were maintained in either the RPMI-1640 medium (Roswell Park Memorial Institute) (PNT1A, LNCaP, LNCaP Mock, LNCaP Hsp27) or DMEM medium (Dulbecco's Modified Eagle's Medium) (PC3, DU145), supplemented 10% fetal bovine serum (FBS) at 37° C. in 5% CO2.

[0194] Human Prostate Cancer TMA Construction and Immunohistochemistry (IHC)

[0195] A total of 352 specimens [33 benign prostate tumours, 94 primary tumours without lymph node metastasis, 61 primary tumours with lymph node metastasis, 78 neo-adjuvant-treated primary tumours, 86 castration-resistant prostate cancers (CRPC)] were collected from the Vancouver Prostate Centre Tissue Bank. The TMA construction has previously been described (Thomas et al. 2011). IHC staining with mouse anti-DDX5 monoclonal antibody (sc-365164, Santa Cruz Biotechnology, CA, USA) was performed using a Ventana autostainer (model Discover XT; Ventana Medical System, Tucson, Ariz.) with an enzyme-labelled biotin-streptavidin system and a solvent-resistant DAB Map kit (Ventana). We quantified the staining of DDX5 on each cores with the Quick Score (QS) which combined the intensity of the staining and the percentage of stained cells. The level of DDX5 immunostaining was scored on a scale from 0 to 3 by pathologist as described (Choi et al. 2016). Basically, 0 was undetectable stain, 1 means a faint stain, 2 means a stain with obvious intensity occurring in a minority of cells, and 3 represents for a stain with convincing intensity in almost of cells.

[0196] Development of DDX5-Targeting ASOs

[0197] Developing DDX5-tageting ASOs has been performing in the “Optoligo” national platform which is sponsored by Inserm Transfert. The sequence of ASOs was designed by using an informatics R script which created by Pascal FINETTI in our laboratory. Firstly, the coding part of DDX5 mRNA sequence (CCDS11659.1) was selected and segmented into consecutive sequences of 20 bases. The resulting sequences were converted into their complementary acid nucleotide sequences, which subsequently were inverted to obtain the 5′ to 3′ format and become potential ASO sequences. In a second step, the ASO sequences were individually evaluated on the percentage of GC and the specificity to the input transcripts. The specificity evaluation was performed by using NCBI's ‘Basic Local Alignment Search Tool’ (BLAST), with the use of the ‘blastn’ algorithm and the NCBI reference transcript database ‘refseq_rna’ as parameters. The next selection was done manually by excluding ASOs showing sequence similarities with other genes. Finally, the selected ASOs were synthesized by Pr. Philippe BARTHELEMY (ARNA Laboratory, Inserm U1212, CNRS UMR5320, University of Bordeaux). The biological activity of the different ASOs was then tested on PC tumor cell line, the inhibition of DDX5 protein was evaluated by Western blot (WB) and then quantified with Image J (NIH) software.

[0198] Treatment Cells with ASOs

[0199] Seeding of the cells was carried out 1 day before treatment at a density of 2300 cells per 1 cm.sup.2. The cells were treated with indicated ASO concentrations after incubating with 3 mg/ml of Oligofactamine (Invitrogen) in serum free OPTI-MEM medium (Invitrogen) for 20 minutes. After 4-5 hours, the transfection mixtures were removed and replaced by the complete medium. One day later, the second treatment was performed identically. The transfected cells were collected after 3 days of the latest ASO transfection. We used scrambled ASO as control.

[0200] Immunoprecipitation Coupled Mass Spectrometry (IP/MS)

[0201] An amount of 2 mg protein extracts from 4 cell lines (PNT1a, LNCaP, DU 145, and PC 3) was diluted with the lysis buffer to obtain the final concentration of 4 mg protein per 1 ml. Subsequently, the protein extracts were pre cleaned with 40 μl of the protein A Sepharose (nProtein A Sepharose® 4 Fast Flow, REF.17-5280-01, GE Healthcare, MERCK). Subsequently, the lysate was incubated with 5 μg of Ab against DDX5 (mouse monoclonal Ab, sc-365164, Santa Cruz Biotechnology) overnight at 4° C. The immunoprecipitated complexes were then captured by incubation with 40 μl of protein A Sepharose bead for 1 hr, 4° C., which was followed by 3 times of washing using the lysis buffer. Ultimately, the resulting beads were suspended with 20 μl Laemmli sample buffer 4×, heated at 95° C. for 5 minutes. To evaluate the efficiency of the IP, 10% of the samples were run on the SDS-PAGE gel for silver staining analysis as described (Chevallet et al. 2006). The proteins in immunoprecipitated complexes were determined by LC-MS/MS using LTQ Orbitrap (Thermoscientific). The IP experiments were done with triplications and 3 technical replications.

[0202] Western Blot Analysis (WB)

[0203] The total protein extracts were obtained by resuspension the cell pellet with the lysis buffer (1% v/v Triton X-100, 50 mM HEPES, 150 mM NaCl, 25 mM NaF, 1 mM EDTA, 1 mM EGTA, 10 μM ZnCl2, 1 mM sodium orthovanadate) containing 4% v/v protease inhibitor cocktail (Roche) and incubated for 20 minutes on ice. After centrifugation at 13 000 rpm, 4° C., 30 minutes, the clear lysate was collected and quantified by using BCA protein assay kit (Pierce). The protein (20 μg per lane) was pre mixed with the Laemmli sample buffer 4× and boiled at 95° C. for 5 minutes before running on the SDS-polyacrylamide gels (10%). The migrated proteins were then transferred into the polyvinylidene difluoride (PVDF) membranes (Millipore). The resulting membranes blocked with 5% w/v nonfat milk in Tris-buffered saline (TBS) and probed with 1:10000 rabbit anti-Hsp27 polyclonal antibody (Enzo Life Science, Villeurbanne, France); 1:800 mouse anti-DDX5 monoclonal antibody (sc-365164), 1:500 mouse anti-Ku86 monoclonal antibody (sc-5280), 1:500 mouse anti-Ku70 monoclonal antibody (sc-17789), 1:500 mouse anti-NF45 monoclonal antibody (sc-365283) (Santa Cruz Biotechnology, CA, USA). Vinculin and GAPDH were used as a loading control. Subsequently, the primary Ab was probed with corresponding HRP-conjugated secondary Abs (DAKAO) and detected using ECL prime Western Blotting detection reagent (RPN2236, GE Healthcare).

[0204] Results

[0205] Elevated DDX5 Protein Expression is Associated with CRPC.

[0206] We examined DDX5 expression by immunostaining of a human prostate TMA. Prostate cancer has significantly higher level of DDX5 expression than benign prostatic hyperplasia (BPH). Among 143 PC specimens, Gleason grade 5 prostate cancer has the strongest DDX5 expression, following by Gleason 4 and 3 (data not shown), implying that DDX5 expression is correlated to the aggressiveness of the disease. Moreover, increased DDX5 protein expression was observed in tumors from patients under prolonged neoadjuvant hormone therapy (>6 months) and CRPC patients (data not shown), demonstrating that overexpressed DDX5 is correlated with the progression of CRPC. DDX5 expression was also found to be associated with the metastatic or the stage of prostate cancer since the staining in tumors with lymph node (LN) metastatic was more intensive than in tumors without LN metastatic, and lower in CRPC (data not shown). In addition, the median recurrence free survival (RFS) was 45.4 months in the moderate or strong group, corresponding to a 52% higher relative risk of recurrence compared to the negative or weak group with the median RFS of 76.6 months, indicating that DDX5 overexpression was correlated to a short recurrence free survival (RFS). On the other hand, when it comes to the comparison of DDX5 protein expression between CS (LNCaP) and the CR PC cells (DU145+PC3) by Western Blot, DDX5 expression levels in CR cells are much higher than in the CS cell. Taken together, these results reinforce the point that DDX5 represents a potential target that is relevant in CRPC.

[0207] Developing ASOs Targeting DDX5

[0208] Elevated DDX5 protein expression is correlated to the PC advancement and CRPC progression, so developing a DDX5 inhibitor could be a valuable approach for CRPC treatment. First, the program designed 93 ASOs against the human mRNA sequence of DDX5 (hASO) (data not shown). The 1st screening tested 13 hASOs and showed that hASO #51 decrease DDX5 expression by 81% at 200 nM and in dose-dependent manner (FIG. 1A). Moreover, DDX5 inhibition induced by hASO #51 resulted in a significant decrease in cell viability AI cell line (DU-145) (FIG. 1B). Prior to testing a DDX5 ASO on xenografted mouse models with our human AI cell lines, we continued the screening of DDX5 ASOs to find out a bi-specific sequence of both human and murine form, allowing us to apprehend potential toxic effects of DDX5 inhibition in tumors-carried mice. The percentage of bi-specificity was examined by BLAST analysis of the human and murine form (hmASO), giving 25 bi-specific sequences (data not shown). We tested 8 sequences distributed uniformly over the entire target sequence at different locations of the mRNA in addition to hASO #51. We found that hmASO #3 depleted DDX5 protein expression as effectively as hASO #51 and in a dose-dependent manner (FIGS. 1C and 1D). Therefore, in order to check if DDX5 inhibition is able to restore the tumour sensitivity to CPRC treatments, the hmASO #3 currently becomes the leading DDX5 ASO that is chosen for our in vivo experiment in mice model.

[0209] DDX5 is an Hsp27 Partner and Hsp27-Regulated Protein

[0210] Hsp27, a small Heat shock protein (sHSP), behaviors as a oncogene during PC progression, its overexpression is positively correlated with metastasis and Castration Resistance emergence (Rocchi et al. 2005). An Antisense Oligonucleotide targeting Hsp27, namely OGX 427, has demonstrated to restore castration and chemotherapy sensitivity of the PC cell. In order to shed light on the mechanism of action in which Hsp27 drives PC initiation and CRPC evolution, several studies which were based on proteomics approaches have been performed in our laboratory. Our previous study aiming to search for the Hsp27- regulated proteins by proteomic profiling comparison between LNCaP-Hsp27 and LNCaP-Mock indicated that DDX5 protein abundance is correlated with Hsp27 protein level (unpublished data). The DDX5 protein level indeed was proved to be higher in the LNCaP-Hsp27 in which Hsp27 gene was stably transfected compared to the LNCaP-Mock cells. Moreover, Hsp27 depletion by OGX-427 decreased significantly DDX5 level in both LNCaP and PC3 (data not shown). Our study on Hsp27 interactome on different PC cell lines using IP/MS suggested that DDX5 is an Hsp27 binding protein (prepared data for submission). The interaction between Hsp27 and DDX5 was confirmed by WB following IP on PC-3 cells (data not shown). All of these results together demonstrated that Hsp27 interacts with DDX5 and regulates DDX5 expression, and Hsp27 might function as a chaperone protein that protects DDX5 from miss folding and ubiquitin-proteasome degradation.

[0211] Through Oncomine meta-analysis, DDX5 found to be co-expression with several proteins belonging to the ubiquitin pathway (USP9X, UBE2JI, and UBE3A) and the proteasome (PSMA2, PRKWNK), suggesting the interaction of DDX5 with these proteins and its involvement in these pathways (Wilson and Giguère 2007). Previously, DDX5 was demonstrated to be a target of poly-ubiquitination, hence its stability might be likely regulated through ubiquitin-proteasome system (Causevic et al. 2001) (Mooney et al. 2010). In order to check if the stability of DDX5 is regulated by the 26S proteasome in prostate cancer, the PC3 cells were treated with MG132 at 10 an inhibitor of the proteasome, and with or without cycloheximide (CHX) 10 μg/ml to inhibit the de novo protein synthesis with indicated point of time. Indeed, the DDX5 level increased over course of time due to a blocked proteasome by MG132 (data not shown), and went down by 50%, 70% after inhibiting of protein synthesis 24 hours, 30 hours, respectively (data not shown); meanwhile, it was conserved over time when the PC 3 cells were treated with both MG132, and cycloheximide (CHX) (data not shown). These all suggest that proteasome is likely the main pathway controls the degradation of DDX5 in PC.

[0212] Our recent studies have demonstrated that Hsp27 interacts with TCTP (Translationally-controlled tumor protein) and eIF4E (eukaryotic translation initiation factor 4E) and protects these protein partners from proteasome degradation (Andrieu et al. 2010) (Baylot et al. 2012) DDX5 protein levels are elevated in many cancers as mentioned above, but not as a result of upgraded mRNA levels (Causevic et al. 2001). Our DNA microarray data on various models which were up regulated or down regulated of Hsp27 protein expression did not show any difference of the DDX5 mRNA levels among different comparisons such as LNCaP-Hsp27 vs LNCaP-OGX427, LNCaP Hsp27 vs LNCaP, PC3_OGX 427 vs PC3_SCR (data not shown). Indeed, DDX5 expression was confirmed to not be controlled at the transcription level by Hsp27 (data not shown), suggesting that Hsp27 may control DDX5 protein stability. To identify how Hsp27 modulates DDX5 abundance, the DDX5 half-life was determined after OGX 427 treatment. The PC3 cells were transfected with ASO of Hsp27, OGX427 or the ASO control and treated with or without MG132/CHX for 48 hours before being harvested for protein extractions. MG132/CHX treatment extents DDX5 half-life and reversed the effect of Hsp27 depletion by ASO OGX-427, suggesting that DDX5 depletion after OGX427-induced Hsp27 knockdown is as a result of proteasome degradation (data not shown)

[0213] In order to know if there is any feedback loop between 2 proteins, we checked the Hsp27 abundance after DDX5 depletion. We did not observe any changes of the Hsp27 protein level due to DDX5 inhibition (data not shown). Therefore, it is likely that Hsp27 controls the DDX5 stability and DDX5 does not regulate Hsp27 expression.

[0214] DDX5 Promotes CRPC by Activating AKT/mTOR1 Pathway

[0215] DDX5 was demonstrated to promote cell survival and growth by activating the mTORC1 signaling pathway (Taniguchi et al. 2016). DDX5 was previously showed to control p-S6K1 and p 4E BP1 levels, two well-known effectors of activated mTOR signaling. We examined if DDX5 can regulate the upstream components belonging to the pathway such as mTOR1, AKT. We found that DDX5 inhibition significantly decreased protein levels of both AKT and p-AKT and p-mTOR (data not shown).

[0216] Identification of DDX5 Interactomes in Various PC Cell Lines by IP Coupled LC-MS/MS

[0217] To shed light on the mechanism of actions by which DDX5 drives CRPC progression, we identified the PPI (protein-protein interaction) networks of DDX5 in four PC cell lines that present for different stages of PC progression (data not shown). The protein complexes of endogenous DDX5 containing its binding proteins were captured by immunoprecipitation (IP) using the antibody against DDX5 and subsequently determined by LS/MS/MS. The software MaxQuant was applied to remove nonspecific binding proteins and give the list of DDX5-associated proteins (data not shown). The silver staining analyses 10% of the IP elution showed the band of DDX5 on the IP samples but not on the controls, which illustrated that our IP experiment works very well (data not shown). Three biological replicates were performed along with 3 technical triplicate of LC-MS/MS and gained very high reproducibility (correlation value <8e-1) for all the cell lines (data not shown). In total, 489 proteins were identified, and we obtained 16, 44,239 and 401 candidates for PNT1a, LNCaP, DU 145, and PC-3 respectively (data not shown). We also classified the proteins found by IP/MS into 2 classes: high evidence (present in both filters FDR 0.1 and FDR=1) and the 2 one is just satisfactory FDR=1 but not FDR=0.1. Obviously, an association of the number of DDX5 binding proteins with the aggressiveness of the disease and CRPC was observed, this indicates that DDX5 might extent more cellular functions during the disease development. The Venny diagram showed a cross among 4 lines ‘datasets (data not shown). 173 candidates are shared between 2 CR lines (DU 145 and PC 3), which covered up to 72.3% of the interactors determined in DU 145 (data not shown). The PPI interaction network of the 489 proteins obtained in four cell lines which were constructed using the STRING database showed a very high number of connections among 487 nodes (487/489, 99.9%) with 13522 edges (average node degree: 55.5, avg. local clustering coefficient: 0.54, PPI enrichment p-value: <1.0e-16). DDX5 is connected to 113 proteins (113/487, approximately 23.2%) in the network, (data not shown). Especially, we found 59/113 proteins inside the network which were considered to have known interactions with DDX5 from both experimentally determined and curated databases. DDX17 which is known as DDX5 paralog, and well-described interactor of DDX5 were found in all of 4 cell line dataset, and in the network with combined score 0.75 (Ogilvie et al. 2003). In addition, we also obtained TP53, DHX9,CDK9, which are very well described as DDX5 partners (Bates et al. 2005) (Nicol et al. 2013a) (Wilson and Giguère 2007) (Yang et al. 2015). A number of proteins have very high combined score (above 0.9) such as HNRNPL, ELAVL1, FUS, HNRNPA0, SRSF1, and YBX1; and so on (data not shown). These all together proved the highly efficiency of our Co-IP-LC-MS/MS.

[0218] Characterization of the Global DDX5 Interactome in PC

[0219] The DDX5-interacting proteins found in 4 cell lines was functional classified by using PANTHER 14.1. When it comes to Molecular function, nearly a half of the DDX5 interactome involve in “binding” in which they interact with other molecules such as: nucleic acid, protein, protein containing complex, and chromatin. A significant number of the DDX5 partners possess enzyme activity (catalytic activity: 22%) and contribute to complex assembly (structural molecule activity: 20%) (data not shown). In the consistent with these, classifying the DDX5 interactome based on the PANTHER “protein class” database showed that a majority of the DDX5 interactors are involved in binding to nucleic acid and most of them are RNA binding proteins (data not shown). It also revealed several protein classes engaging with catalytic functions such as hydrolase (9.2%), transferase (8.1%), ligase (1.4%), and enzyme modulator (4.6%). Interestingly, we could recognize a set of DDX5-associated proteins which function as transcription factors (23 proteins, 6.6%) and 8 of them are zinc finger transcription factors.

[0220] In order to decipher the putative functions associated with DDX5-interacting proteins in PC, we performed the analysis of Gene Ontology (GO) enrichment using BiNGO tool run by Cytoscape (Maere et al. 2005) for the set of total DDX5 interactors found in 4 cell lines, and 126, and 79 functions were obtained with significant level p=0.05 and 0.005, respectively (data not shown). As expected, well-known key functions of DDX5 were observed. We found that DDX5 plays vital roles in gene expression since a majority of DDX5 interactors (up to 57%) are engaged with this process (p=0.0000E-100). DDX5 functions mainly in RNA processing (p=1.2751E-72), translation (p=1.8605E-71), ribosome biogenesis (p=1.8027E-42) and transcription (p=2.8988E-6). In addition, DDX5 likely modulates gene expression at posttranscriptional levels by regulating mRNA stability, RNA splicing and translation. On the other hand, we found that DDX5 participates in the cellular response to the DNA damage stimulus (p=4.4587E-3) including DNA repair process. Beside involving to the p53 related-DNA damage response as described before (Nicol et al. 2013b), we found novel functions of DDX5 in DNA repair pathways such as: Non homologous end-joining (NHEJ) and Nucleotide excision repair (NER).

[0221] By performing a functional enrichment analysis using Gprofiler (Raudvere et al. 2019) for our DDX5 associated proteins with the CORUM 3.0 database (https://mips.helmholtz-muenchen.de/corum/) which collects annotation of mammalian protein complexes obtained from manual experiments (Giurgiu et al. 2019), we determined various protein complexes corresponding to different functions in the DDX5 interactome (data not shown). Consistently, we found that DDX5 tightly associates with the complexes involved in ribosome biogenesis, protein synthesis in cytoplasm and mitochondria, splicing, mRNA stability, transcription and DNA repair.

[0222] DDX5 likely modulates the mRNA stability via interacting with IGF2BPs complex which is consisted of 9 proteins: RPS6, RPL26, DHX9, STAU1, ELAVL1, SYNCRIP, HNRNPU, IGF2BPs (IGF2BP1,IGF2BP2, IGF2BP3), and YBX1 (Weidensdorfer et al. 2009). The main function of IGF2BPs complex is to enhance the stability and storage of the target mRNA by associating with the Coding Region instability Determinant (CRD) (Huang et al. 2018), including IGF2 (Dai et al. 2017) (Cao et al. 2018), MYC (Noubissi et al. 2006) (Weidensdorfer et al. 2009), ACTIN (Hüttelmaier et al. 2005) and LIN28B (Hafner et al. 2010). Moreover, DHX9, IGF2BPs and YBX1 have been proved to play oncogenic function and therapy resistant in various cancers. DDX5 tightly associates with the IGF2BP2 and IGF2BP3 complex since we found all of their protein components in our DDX5 interactome (data not shown). Moreover, DDX5 has been showed to be associated with DHX9, STAU1, ELAVL1, SYNCRIP, HNRNPU, IGF2BP3, and YBX1 based on string database.

[0223] The identified DDX5 interactome showed a solid connection between DDX5 and toposome, revealing novel mechanism of action by which DDX5 regulates cell cycle. Toposome which consists of TOP2A, SRPK1, DHX9, HNRNPC, PRPF8, DDX21, and SSRP1 plays essential roles in cell cycle regulation by modeling chromosome segregation, chromosome topological changes (Lee et al. 2004). Except SSRP1 found to interact with DDX5 in database, all of proteins belonging to Toposome were determined to associate with DDX5 in our study. We first introduced the interaction of DDX5 and toposome, especially TOP2A, a widespread drug target in many types of cancer (Nitiss 2009) (Pogorelcnik and Solmajer 2013). Moreover, our study also confirmed the interaction between DDX5 and TOP1, which was annotated in the string database. This can provide a clue about the mechanism by which DDX5 overexpression conferred to resistance to Camptothecin (CPT), the inhibitor of TOP1 (Cohen et al. 2008).

[0224] Interestingly, we first showed the rigid association of DDX5 with the transcription factor complex TH2H which is composed of the core complex (GTF2H1, GTF2H2, GTF2H3, GTF2H4, ERCC2, ERCC3) and the CAK (CDK7, CCNH, MNAT1) (data not shown). The TH2H complex functions in both transcription and DNA damage response, suggesting the involvement of DDX5 in these biological processes.

[0225] Another novel mechanism in which DDX5 modulates transcription is through the 7SK RPN complex. All the proteins constructing to the complex are found to associate with DDX5 in our study, including CDK9, HEXIM1, CCNT1, and LARP7. The positive transcription elongation factor B, p-TEFb which is composed of CDK9 and CCNT1 is tightly inactivated by its association with LARP7, HEXIM1 and 7SK. Activated P-TEFb is recruited to the initiation complex by interacting with Transcription factors and phosphorylates CTD Ser2 for effective elongation (Romano 2013) (Rahaman et al. 2016).Especially, p-TEFb was shown to be activated by PSA eRNA by which it regulates AR targeted genes transcription in CRPC, promoting Castration resistance emergency (Zhao et al. 2016). The significant of the association between DDX5 and the 7SK complex and p-TDFb is uncharacterized, it calls for further investigation.

[0226] DDX5 Promotes CRPC by Regulation of DNA Damage Response

[0227] By using the ClusterProfiler R packages (Yu et al. 2012) we compared the Gene Ontology Biological Functions (GO BP) among different DDX5 interactomes found in NM, CS, CR cells. The results revealed that DDX5 involves in much more biological functions in CR cells, which mainly related to DNA damage response, translation, transcription, RNA stability, and DNA conformation changes (data not shown). It is believed that DDX5 can promote CRPC development by its participation in these set of exclusive functions found in CR cells.

[0228] It worth noting that we determined a number of novel DDX5 binding proteins involved in DNA repair and exclusively found in the CR lines, uncovering new potential roles of DDX5 in DNA damage response, genomic integrity and CRPC development. The proteins belonging to DNA repair function relied on the BinGO analyses in DU 145 and PC 3 cells are 16 and 15, respectively (data not shown). The main common DNA repair pathways found in both CR cell lines are NHEJ (NHEJ-dependent DSBR) and NER (data not shown). Visualization of these two pathways by KEGG database via Pathview by R showed the DDX5 interactors involved in NHEJ and NER mechanisms (data not shown). The PPI network of DDX5 and 20 proteins belonging to DNA repair module was generated by STRING, and DDX5 was showed to have known interaction with 3 proteins: P53, UPF1, and PRP19. This means that our approach allowed us to discover novel DDX5 binding proteins which participate in DDR, revealing new potential functions of DDX5 in two DNA repair pathways, NHEJ (via Ku complex) and NER (via GTF2H complex).

[0229] Our IP/MS analyses indicated that DDX5 is associated with the Ku70/ku86 complex. We confirmed the binding of DDX5 and the Ku70/ku86 complex by WB followed IP. Ku 70, Ku 86 and NF45 (ILF2) present in the IP using the Ab against DDX5 on both cell lines DU 145 and PC 3 (data not shown). On the other hand, DDX5 is found in the Reverse IP (RIP) with Ku70 and Ku86 Ab (data not shown). These prove the interaction of DDX5 with the core complex of NHEJ, Ku70/Ku86.

[0230] To shed light on DDX5 potential functions in DNA damage repair, we examined how knock out DDX5 affects recovery of DNA repair. The DU 145 cells were transfected with SCR-ASO as a control and ASO51 to induce DDX5 depletion, and objected to irradiation after 72h of transfection. The dynamic of DNA damage recovery was analyzed by IF using anti-p gH2AX foci over time-points: 0 hr, 0.25 hr, 1 hr, 3 hr, 6h, and 24 hr. The foci numbers in the DDX5-depleted cells were lower than those in the control samples from 3 hrs to 24 hrs (data not shown). This implied that DDX5 knock down enhanced efficiency of DNA repair recovery. In another words, DDX5 negatively regulated DNA damage repair.

[0231] ASO3 Downregulates DDX5 Protein Expression Highly Effectively.

[0232] Different batches of ASO3 and ASO51 was then tested for their ability to downregulate DDX5 expression. The results confirmed that ASO3 downregulates DDX5 protein expression highly effectively and even better than the ASO51 in DU-145 cells (FIG. 2). While using ASO3 at 150 nM induced around 80% DDX5 inhibition (FIG. 2), it caused too much death cells making it difficult to obtain protein extract. Basing on this results, ASO3 and nearby ASO3 at 120 nM are tested. The results showed that ASO3-1, ASO3+1 and ASO3+3 also inhibit DDX5 protein level beside the ASO3 (FIG. 3). From all these results, it is concluded that ASO3 is the one which can inhibit DDX5 expression more effectively than other tested ASOs in DU-145 cells.

CONCLUSION

[0233] DDX5 has previously described to be involved in DNA damage response (Nicol et al. 2013b). In this publication, DDX5 was shown to recruit both p53 and RNAPII to the p21 promoter upon irradiation stress, resulting in cell cycle arrest after DNA damage. The study of the inventors has demonstrated novel mechanism of actions of DDX5 involving in DNA damage response in prostate cancer. DDX5 is likely the central of different DNA repair pathways, such as NHEJ, NER. They have proved that depleted DDX5 enhances significantly DNA damage recovery. In another words, DDX5 negatively regulates DNA repair process, providing more advantages for the survival of the tumor cells upon DNA damage-caused stresses such as chemotherapy or irradiation.

REFERENCES

[0234] Throughout this application, various references describe the state of the art to which this invention pertains. The disclosures of these references are hereby incorporated by reference into the present disclosure. [0235] Andrieu C, Taieb D, Baylot V, et al (2010) Heat shock protein 27 confers resistance to androgen ablation and chemotherapy in prostate cancer cells through eIF4E. Oncogene 29:1883-1896. https://doi.org/10.1038/onc.2009.479 [0236] Bates G J, Nicol S M, Wilson B J, et al (2005) The DEAD box protein p68: a novel transcriptional coactivator of the p53 tumour suppressor. EMBO J 24:543-553. https://doi.org/10.1038/sj.emboj.7600550 [0237] Baylot V, Katsogiannou M, Andrieu C, et al (2012) Targeting TCTP as a new therapeutic strategy in castration-resistant prostate cancer. Mol Ther J Am Soc Gene Ther 20:2244-2256. https://doi.org/10.1038/mt.2012.155 [0238] Bond A T, Mangus D A, He F, Jacobson A (2001) Absence of Dbp2p alters both nonsense-mediated mRNA decay and rRNA processing. Mol Cell Biol 21:7366-7379. https://doi.org/10.1128/MCB.21.21.7366-7379.2001 [0239] Cao J, Mu Q, Huang H (2018) The Roles of Insulin-Like Growth Factor 2 mRNA-Binding Protein 2 in Cancer and Cancer Stem Cells. Stem Cells Int 2018: https://doi.org/10.1155/2018/4217259 [0240] Caretti G, Schiltz R L, Dilworth F J, et al (2006) The RNA helicases p68/p72 and the noncoding RNA SRA are coregulators of MyoD and skeletal muscle differentiation. Dev Cell 11:547-560. https://doi.org/10.1016/j.devce1.2006.08.003 [0241] Causevic M, Hislop R G, Kernohan N M, et al (2001) Overexpression and poly-ubiquitylation of the DEAD-box RNA helicase p68 in colorectal tumours. Oncogene 20:7734. https://doi.org/10.1038/sj.onc.1204976 [0242] Chevallet M, Luche S, Rabilloud T (2006) Silver staining of proteins in polyacrylamide gels. Nat Protoc 1:1852-1858. https://doi.org/10.1038/nprot.2006.288 [0243] Chi K N, Yu E Y, Jacobs C, et al (2016) A phase I dose-escalation study of apatorsen (OGX-427), an antisense inhibitor targeting heat shock protein 27 (Hsp27), in patients with castration-resistant prostate cancer and other advanced cancers. Ann Oncol 27:1116-1122. https://doi.org/10.1093/annonc/mdw068 [0244] Choi S Y C, Xue H, Wu R, et al (2016) The MCT4 Gene: A Novel, Potential Target for Therapy of Advanced Prostate Cancer. Clin Cancer Res Off J Am Assoc Cancer Res 22:2721-2733. https://doi.org/10.1158/1078-0432.CCR-15-1624 [0245] Clark E L, Coulson A, Dalgliesh C, et al (2008) The RNA helicase p68 is a novel androgen receptor coactivator involved in splicing and is overexpressed in prostate cancer. Cancer Res 68:7938-7946. https://doi.org/10.1158/0008-5472.CAN-08-0932 [0246] Clark E L, Hadjimichael C, Temperley R, et al (2013) p68/DdX5 Supports β-Catenin & RNAP II during Androgen Receptor Mediated Transcription in Prostate Cancer. PLOS ONE 8:e54150. https://doi.org/10.1371/journal.pone.0054150 [0247] Cohen A A, Geva-Zatorsky N, Eden E, et al (2008) Dynamic proteomics of individual cancer cells in response to a drug. Science 322:1511-1516. https://doi.org/10.1126/science.1160165 [0248] Dai N, Ji F, Wright J, et al IGF2 mRNA binding protein-2 is a tumor promoter that drives cancer proliferation through its client mRNAs IGF2 and HMGA1 eLife 6: https://doi.org/10.7554/eLife.27155 [0249] Dardenne E, Pierredon S, Driouch K, et al (2012) Splicing switch of an epigenetic regulator by RNA helicases promotes tumor-cell invasiveness. Nat Struct Mol Biol 19:1139-1146. https://doi.org/10.1038/nsmb.2390 [0250] Fuller-Pace F V (2006) DExD/H box RNA helicases: multifunctional proteins with important roles in transcriptional regulation. Nucleic Acids Res 34:4206-4215. https://doi.org/10.1093/nar/gkl460 [0251] Guturi K K N, Sarkar M, Bhowmik A, et al (2014) DEAD-box protein p68 is regulated by β-catenin/transcription factor 4 to maintain a positive feedback loop in control of breast cancer progression. Breast Cancer Res BCR 16: https://doi.org/10.1186/s13058-014-0496-5 [0252] Hafner M, Landthaler M, Burger L, et al (2010) Transcriptome-wide identification of RNA-binding protein and microRNA target sites by PAR-CLIP. Cell 141:129-141. https://doi.org/10.1016/j.ce11.2010.03.009 [0253] Huang H, Weng H, Sun W, et al (2018) Recognition of RNA N6-methyladenosine by IGF2BP proteins enhances mRNA stability and translation. Nat Cell Biol 20:285-295. https://doi.org/10.1038/s41556-018-0045-z [0254] Hüttelmaier S, Zenklusen D, Lederer M, et al (2005) Spatial regulation of beta-actin translation by Src-dependent phosphorylation of ZBP1. Nature 438:512-515. https://doi.org/10.1038/nature04115 [0255] Jalal C, Uhlmann-Schiffler H, Stahl H (2007) Redundant role of DEAD box proteins p68 (Ddx5) and p72/p82 (Ddx17) in ribosome biogenesis and cell proliferation. Nucleic Acids Res 35:3590-3601. https://doi.org/10.1093/nar/gkm058 [0256] Janknecht R (2010) Multi-talented DEAD-box proteins and potential tumor promoters: p68 RNA helicase (DDX5) and its paralog, p72 RNA helicase (DDX17). Am J Transl Res 2:223-234 [0257] Karaki S, Paris C, Rocchi P (2019) Antisense Oligonucleotides, A Novel Developing Targeting Therapy. Antisense Ther. https://doi.org/10.5772/intechopen.82105 [0258] Katsogiannou M, Ziouziou H, Karaki S, et al (2015) The hallmarks of castration-resistant prostate cancers. Cancer Treat Rev 41:588-597. https://doi.org/10.1016/j.ctrv.2015.05.003 [0259] Lee C-G, Hague L K, Li H, Donnelly R (2004) Identification of toposome, a novel multisubunit complex containing topoisomerase IIalpha. Cell Cycle Georget Tex 3:638-647 [0260] Lin S, Tian L, Shen H, et al (2013) DDX5 is a positive regulator of oncogenic NOTCH1 signaling in T cell acute lymphoblastic leukemia. Oncogene 32:4845-4853. https://doi.org/10.1038/onc.2012.482 [0261] Ma Z, Feng J, Guo Y, et al (2017) Knockdown of DDX5 Inhibits the Proliferation and Tumorigenesis in Esophageal Cancer. Oncol Res 25:887-895. https://doi.org/10.3727/096504016X14817158982636 [0262] Mazurek A, Luo W, Krasnitz A, et al (2012) DDX5 regulates DNA replication and is required for cell proliferation in a subset of breast cancer cells. Cancer Discov 2:812-825. https://doi.org/10.1158/2159-8290.CD-12-0116 [0263] Mooney S M, Grande J P, Salisbury J L, Janknecht R (2010) Sumoylation of p68 and p72 RNA helicases affects protein stability and transactivation potential. Biochemistry 49:1-10. https://doi.org/10.1021/bi901263m [0264] Nicol S M, Bray S E, Black H D, et al (2013a) The RNA helicase p68 (DDX5) is selectively required for the induction of p53-dependent p21 expression and cell-cycle arrest after DNA damage. Oncogene 32:3461-3469. https://doi.org/10.1038/onc.2012.426 [0265] Nicol S M, Bray S E, Black H D, et al (2013b) The RNA helicase p68 (DDX5) is selectively required for the induction of p53-dependent p21 expression and cell cycle arrest after DNA damage. Oncogene 32:3461-3469. https://doi.org/10.1038/onc.2012.426 [0266] Nyamao R M, Wu J, Yu L, et al (2019) Roles of DDX5 in the tumorigenesis, proliferation, differentiation, metastasis and pathway regulation of human malignancies. Biochim Biophys Acta Rev Cancer 1871:85-98. https://doi.org/10.1016/j.bbcan.2018.11.003 [0267] Ogilvie V C, Wilson B J, Nicol S M, et al (2003) The highly related DEAD box RNA helicases p68 and p72 exist as heterodimers in cells. Nucleic Acids Res 31:1470-1480. https://doi.org/10.1093/nar/gkg236 [0268] Pogorel_cnik B, Solmajer A P and T (2013) Recent Advances in the Development of Catalytic Inhibitors of Human DNA Topoisomerase IIα_As Novel Anticancer Agents. In: Curr. Med. Chem. http://www.eurekaselect.com/106834/article. Accessed 7 Oct. 2019 [0269] Rahaman M H, Kumarasiri M, Mekonnen L B, et al (2016) Targeting CDK9: a promising therapeutic opportunity in prostate cancer. Endocr Relat Cancer 23:T211-T226. https://doi.org/10.1530/ERC-16-0299 [0270] Rinaldi C, Wood M J A (2018) Antisense oligonucleotides: the next frontier for treatment of neurological disorders. Nat Rev Neurol 14:9-21. https://doi.org/10.1038/nrneuro1.2017.148 [0271] Rocchi P, Beraldi E, Ettinger S, et al (2005) Increased Hsp27 after Androgen Ablation Facilitates Androgen-Independent Progression in Prostate Cancer via Signal Transducers and Activators of Transcription 3-Mediated Suppression of Apoptosis. Cancer Res 65:11083-11093. https://doi.org/10.1158/0008-5472.CAN-05-1840 [0272] Rocchi P, Jugpal P, So A, et al (2006) Small interference RNA targeting heat-shock protein 27 inhibits the growth of prostatic cell lines and induces apoptosis via caspase-3 activation in vitro. BJU Int 98:1082-1089. https://doi.org/10.1111/j.1464-410X.2006.06425.x [0273] Rocchi P, So A, Kojima S, et al (2004) Heat shock protein 27 increases after androgen ablation and plays a cytoprotective role in hormone-refractory prostate cancer. Cancer Res 64:6595-6602. https://doi.org/10.1158/0008-5472.CAN-03-3998 [0274] Romano G (2013) Deregulations in the Cyclin-Dependent Kinase-9-Related Pathway in Cancer: Implications for Drug Discovery and Development. In: Int. Sch. Res. Not. https://www.hindawi.com/journals/isrn/2013/305371/. Accessed 27 Jun. 2019 [0275] Rosenberg J E, Hahn N M, Regan M M, et al (2018) Apatorsen plus docetaxel versus docetaxel alone in platinum-resistant metastatic urothelial carcinoma (Borealis-2). Br J Cancer 118:1434-1441. https://doi.org/10.1038/s41416-018-0087-9 [0276] Shin S, Rossow K L, Grande J P, Janknecht R (2007) Involvement of RNA helicases p68 and p72 in colon cancer. Cancer Res 67:7572-7578. https://doi.org/10.1158/0008-5472.CAN-06-4652 [0277] Taniguchi T, Iizumi Y, Watanabe M, et al (2016) Resveratrol directly targets DDX5 resulting in suppression of the mTORC1 pathway in prostate cancer. Cell Death Dis 7:e2211. https://doi.org/10.1038/cddis.2016.114 [0278] Thomas C, Zoubeidi A, Kuruma H, et al (2011) Transcription factor StatS knockdown enhances androgen receptor degradation and delays castration-resistant prostate cancer progression in vivo. Mol Cancer Ther 10:347-359. https://doi.org/10.1158/1535-7163.MCT-10-0850 [0279] Wang Z, Luo Z, Zhou L, et al (2015) DDX5 promotes proliferation and tumorigenesis of non-small-cell lung cancer cells by activating β-catenin signaling pathway. Cancer Sci 106:1303-1312. https://doi.org/10.1111/cas.12755 [0280] Weidensdorfer D, Stohr N, Baude A, et al (2009) Control of c-myc mRNA stability by IGF2BP1-associated cytoplasmic RNPs. RNA N Y N 15:104-115. https://doi.org/10.1261/rna.1175909 [0281] Wilson B J, Giguère V (2007) Identification of novel pathway partners of p68 and p72 RNA helicases through Oncomine meta-analysis. BMC Genomics 8:419. https://doi.org/10.1186/1471-2164-8-419 [0282] Yang J, Zhao Y, Kalita M, et al (2015) Systematic Determination of Human Cyclin Dependent Kinase (CDK)-9 Interactome Identifies Novel Functions in RNA Splicing Mediated by the DEAD Box (DDX)-5/17 RNA Helicases. Mol Cell Proteomics 14:2701-2721. https://doi.org/10.1074/mcp.M115.049221 [0283] Yang L, Lin C, Liu Z-R (2006) P68 RNA helicase mediates PDGF-induced epithelial mesenchymal transition by displacing Axin from beta-catenin. Cell 127:139-155. https://doi. org/10.1016/j. cell.2006.08.036 [0284] Yu E Y, Ellard S L, Hotte S J, et al (2018) A randomized phase 2 study of a HSP27 targeting antisense, apatorsen with prednisone versus prednisone alone, in patients with metastatic castration resistant prostate cancer. Invest New Drugs 36:278-287. https://doi.org/10.1007/s10637-017-0553-x [0285] Yu G, Wang L-G, Han Y, He Q-Y (2012) clusterProfiler: an R Package for Comparing Biological Themes Among Gene Clusters. OMICS J Integr Biol 16:284-287. https://doi.org/10.1089/omi.2011.0118 [0286] Zhao Y, Wang L, Ren S, et al (2016) Activation of P-TEFb by Androgen Receptor-Regulated Enhancer RNAs in Castration-Resistant Prostate Cancer. Cell Rep 15:599-610. https://doi.org/10.1016/j.celrep.2016.03.038