DDX20 interlinking with SNRPE endows MYC-driven triple-negative breast cancer therapeutic vulnerability in a splicing-independent manner

doi:10.21203/rs.3.rs-2375888/v1

Download PDF

Research Article

DDX20 interlinking with SNRPE endows MYC-driven triple-negative breast cancer therapeutic vulnerability in a splicing-independent manner

https://doi.org/10.21203/rs.3.rs-2375888/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background: Aberrant splicing is a hallmark of cancer. For MYC-driven cancers, such as triple-negative breast cancer, hypersensitivity to spliceosome-targeted therapy is a novel anticancer strategy since it also triggers antitumorimmunity. Of the approximately 200 known spliceosomal genes, core spliceosomes are essential for cancer cell survival, yet the underlying mechanisms remain elusive.

Methods: First, SNPRE representing the core spliceosomal Sm gene was screened using a multiomics database analysis approach. Then, we investigated the effect of SNRPE on the growth of TNBC cells in vitro and in vivo. We identified DDX20 from the interacting partners of SNPRE as a potential molecule involved in spliceosome vulnerability using coimmunoprecipitationtechniques. The mechanism of the SNRPE-DDX20 axis in tumor spliceosome therapeutic vulnerability was determined by various techniques,including RNA-seq, chromatin immunoprecipitation and other technologies.

Results: In our research, SNRPE was upregulated in TNBC and predicted poor prognosis. Downregulation of SNRPE attenuated the vitality of TNBC in vivo and in vitro. We hypothesized and confirmed DDX20 as an important factor in core spliceosome therapeutic vulnerability. In DDX20-knockout TNBC cells, cell survival is no longer dependent onintracellular splicing activity orSNPRE expression. In DDX20/SNRPE double-downregulated cells, growth ability was restored due to MTOR reactivation via a splicing-independent pathway. In SNRPE-knockdown TNBC cells, ETV3 and its corepressor DDX20 function together to block MTOR expression to induce growth suppression. Once SNRPE expression was high, ETV3 bound to the C-terminus of DDX20 was replaced by SNRPE-containing components, whilethe binding affinity of released ETV3 to MTOR was reduced, and eventually, MTOR was expressed sufficiently to promote proliferation.

Conclusions: Our results reveal that DDX20 is an adaptor in the crosstalk between RNA splicing (SnRNP components such as SNRPE) and the MTOR pathway in MYC-driven TNBC. Patients with high DDX20 expression are likely suitable for treatment with spliceosome modulators, leading to double killing effects by cell arrest and antitumorimmunity. Targeting DDX20 may serve as a therapeutic target for MYC-driven cancers.

RNA splicing is the process of removing noncoding introns from RNA transcripts in eukaryotes and is orchestrated by spliceosomes [1]. The spliceosomes are composed of five small nuclear ribonucleoproteins (SnRNPs) and more than 100 variable associated protein factors, and each SnRNP contains a heptamer of seven different Sm proteins (B/B', D1, D2, D3, E, F, G), which catalyzes both constitutive and alternative splicing [2].

Human cancers exhibit abnormal splicing patterns, and the spliceosome may be a potential “Achilles heel” for MYC-driven cancers [3]. Triple-negative breast cancer (TNBC) is a typical cancer with high MYC expression (gain more than 80%) [4]. In previous years, genome-scale loss-of-function screens of genes essential for the survival of triple-negative breast cancer (TNBC) were annotated as protein degradation (including proteasomal and autophagy-related genes) or spliceosomes [5, 6]. In 72 breast cancer cell lines, screening for survival-essential genes was also focused on the core spliceosomal genes, which were more highly significant in TNBC [7]. In other cancers, survival-essential genes were also the seven Sm genes of the core spliceosomal components with tumor-selective lethal effects [8].

In recent years, spliceosome-targeted therapies (STTs) have the potential for therapeutic vulnerability in MYC-driven cancers [7, 9]. In TNBC, STTs causing intron retention produce neoantigens to trigger antitumor immunity, thereby enhancing immune checkpoint therapy [10, 11]. STTs can also mis-splice some endogenous double-stranded RNAs to induce antiviral immune responses and promote tumor cell death [9].

Once STTs target core spliceosomes, we need to know which cancer patients can elicit a good response of double killing by both cancer-selective growth arrest and anticancer immunity. In this study, we did not perform studies on immunological aspects but focused on exploring the mechanisms of why spliceosome-targeted therapy causes cancer-selective death in cancer cells. Here, we screened the core spliceosomal component SNRPE as a representative of spliceosome-targeted genes. SNRPE downregulation can induce a cancer-selective killing effect. From the SNRPE interacting partners, we selected DDX20 as a key molecule for exploring the mechanism of tumor-selective lethality. Previously, DDX20, also known as GEMIN3 or DP103, was shown to be a multifunctional protein molecule whose primary function is to act as a core component of the SMN complex to participate in the processing of spliceosome biosynthesis, and DDX20 is highly expressed in TNBC and interacts with TAK1 to promote multiorgan tumor metastasis via downstream MMP9 [12]. The function of DDX20 involvement in other cancers has been inconsistently reported [13–16].

Originally, we hope to elucidate the oncogenic mechanism of the DDX20-SNRPE axis from an RNA splicing-related perspective. In TNBC cells, both DDX20 and SNRPE are highly expressed, and downregulation of either SNPRE or DDX20 significantly inhibits splicing ability. The splicing capacity of double-downregulated cells was even more dramatically reduced, but interestingly, tumor growth capacity was unexpectedly restored, implying that the mechanistic effects of the DDX20-SNRPE axis on tumor growth are independent of RNA splicing. Ultimately, in MYC-driven TNBC, we found that DDX20 acts as an adapter to directly link the core pre-mRNA splicing component (SNRPE) and the MTOR signaling pathway. Cancer patients with high DDX20 expression may be suitable for treatment with spliceosome modulators to achieve a dual killing effect through cell growth arrest and antitumor immunity. Targeting DDX20 could be a potential therapeutic target for MYC-driven cancers.

Cell lines and cell culture. Human breast cancer cell lines (MDA-MB-231, MDA-MB-468, HCC-70, MCF-7, T47D, SK-BR-3) and HEK293T cells were originally from the American Type Culture Collection. Luciferase-labeled LM2-4175, a lung metastatic derivative of MDA-MB-231, was kindly provided by Prof. Guo-Hong Hu (University of Chinese Academy of Sciences, Shanghai, China). These human cell lines were cultured in DMEM (Thermo Fisher Scientific) containing 10% fetal bovine serum (Thermo Fisher Scientific) and 1% penicillin/streptomycin (Thermo Fisher Scientific). Two immortalized human mammary epithelial cell lines, HMEC and MCF-10A, were cultured in DMEM/F12 medium supplemented with 5% horse serum, 20 ng/mL epidermal growth factor, 10 µg/mL insulin, 100 ng/mL cholera toxin and 0.5 µg/mL hydrocortis.

Quantitative reverse transcription-PCR (qRT‒PCR). Total RNA was isolated from cell lines using the RNeasy Plus Mini Kit (QIAGEN) and subjected to reverse transcription to cDNA using the PrimeScript RT Reagent Kit with gDNA Eraser (TAKARA). Quantitative PCR was carried out using SYBR Premix Ex Taq (TAKARA) on an ABI 7900HT PCR system (Applied Biosystems). The raw data were processed with SDS v2.1 software in a relative quantification model (-ΔΔC(T)).

Western blot. Immunodetection of each specific protein was carried out according to a standard protocol. Briefly, cultured cells were scraped with lysis buffer (1% SDS, 50 mM Tris-HCl, pH 8.0, 50 mM DTT, protease inhibitor cocktail (Beyotime, China), phosphatase inhibitor cocktail (Roche)) and separated by 8–12% SDS‒PAGE and blotted onto 0.22 µm or 0.45 µm PVDF membranes. The membrane was blocked with 5% nonfat milk in TBS-T buffer and then incubated with specific primary antibody 1:1000 or other dilutions in QuickBlock Primary Antibody Dilution Buffer (Beyotime Biotechnology, China), followed by corresponding HRP-conjugated secondary antibody. After development with SuperSignal West Femto Chemiluminescent Substrate (Thermo Fisher Scientific), images were obtained with a LAS-4000 mini instrument. The following antibodies were used: anti-SNRPE (Proteintech Group, #20407-1-AP), anti-DDX20 (Proteintech Group, #11324-1-AP), anti-ETV3 (Abcam, #ab176717), anti-ACTB (CWbio, China, #CW0098), anti-GAPDH (CWbio, China, # CW0100), anti-Flag (CWbio, China, # CW0287), anti-4E-BP1 (CST, #9452), anti-phospho-4E-BP1 (Ser65) (CST, #9456), and anti-mTOR (7C10) (CST, #2983).

3' rapid amplification of cDNA ends (3’-RACE). The 3’-RACE assay was performed using the SMARTer RACE cDNA Amplification Kit (Clontech Laboratories). Briefly, total RNA from MDA-MB-231 cells was extracted with a RNeasy Mini Kit (QIAGEN), and genomic DNA contamination was removed by on-column RNase-free deoxyribonuclease I (New England Biolabs, USA). The 3'-RACE-ready cDNA of MDA-MB-231 cells was amplified with a SNPRE gene-specific primer and a nested primer GSP with corresponding special oligo (dT) primers. The products were cloned into the pGEM-T easy vector (Promega) for Sanger sequencing.

SNRPE overexpression construct. The CDS of the full-length SNRPE gene (NM_003094) with the Flag sequence and Kozak tag attached at the N-terminal end was synthesized by the Synbio-tech company (Suzhou, China) and cloned into the pcDNA3.1(+) vector at the NheI/BamHI sites. pcDNA3.1-Flag-HA (Addgene #52535) was used as a negative control.

siRNA transfection. Small interfering RNA (siRNA) duplexes with a 3’-terminal dTdT overhang that target human DDX20 or SNRPE mRNA were designed using BLOCK-iT RNAi Designer software (Thermo Fisher Scientific) and synthesized by RiboBio Co., Ltd. (Gungzhou, China). MDA-MB-231 cells were placed in 6-well plates and transfected with 50 nM siRNAs on the following day using Lipofectamine RNAiMAX Reagent (Thermo Fisher Scientific) following the manufacturer's instructions. At 48 h after transfection, the cells were subjected to immunoblotting or other assays.

shRNA plasmid construction. For shRNA vector construction, two validated sequences with high knockdown efficiency that target human SNPRE mRNA were chosen from the Broad TRC lentiviral shRNA library. The annealed synthesized oligonucleotides were ligated into the pLKO.1 plasmid using EcoRI/AgeI restriction enzymes. A human genome nontargeting sequence served as a negative control, and its target sequence was 5’-CAACAAGATGAAGAGCACCAA-3’.

Construction of tetracycline-activated (tet-on) inducible plasmid. The SNRPE shRNA pLKO.1 #1 target sequence was subcloned into a GV307 vector (a pTRIPZ-derived vector, Genechem, China) by digestion with XhoI and EcoRI restriction enzymes and ligation with T4 DNA ligase (New England Biolabs, USA).

Lentivirus packaging and cell infection. For lentivirus particle production, 1.5×106 293FT cells were cultured in 10-cm dishes overnight and cotransfected with 6.0 µg recombinant vector (shRNA or tet-on shRNA) and 1.5 µg pMD2. G and 4.5 µg pspAX2 using Lipofectamine 2000. At 48 h posttransfection, virus-containing supernatant was collected, filtered (0.45 µm filter), aliquoted and stored at -80 ℃ until use. The cells were infected with virus supernatant with 6 µg/ml polybrene for 8 h and selected with puromycin for more than 3 days.

SNRPE inducible knockdown cells. MDA-MB-231 or LM2-4175 cells were infected with shSNRPE tet-on lentivirus, followed by 1.0 µg/ml puromycin (Thermo Fisher Scientific) selection in DMEM containing 10% tetracycline-free FBS for 3 days. To induce RNA interference with SNPRE, gradient concentrations of doxycycline (Sigma) were added to the culture medium for 3 days. Fluorescence microscopy observes TurboRFP fluorescence to verify that the induction is working. Anti-SNPRE immunoblotting to optimize the optimal induction concentration of doxycycline. The puromycin-resistant cells were expanded for subsequent in vivo assays.

Cell proliferation assay. A total of 1 x 10³ or 2 x 10³ cells in 150 µL DMEM were seeded in 96-well plates and imaged using the IncuCyte ZOOM system (Essen BioScience). Frames were captured at 12 h or 24 h intervals from four separate regions per well, and the growth rate was determined using IncuCyte software (2013A Rev2).

SILAC-based identification of interacting partners for SNRPE. 293T cells were cultured in SILAC DMEM (Thermo Fisher Scientific) (supplemented with 10% dialyzed FBS), where half of the 293T cells were labeled with heavy media containing 42 mg/L lysine (U-13C6, 99%; U-15N2, 99%) (Lys8) and 73 mg/L arginine (U-13C6, 99%; U-15N4, 99%) (Arg10) and the other half were labeled with light media containing unlabeled amino acids purchased from Sigma‒Aldrich. Then, 200 mg/L proline (Sigma‒Aldrich) was added to prevent the arginine to proline conversion. After seven passages in the corresponding media, the heavy or light amino acid-labeled cells were transfected with pcDNA3.1-Flag-SNRPE plasmid or control plasmid, respectively, using Lipofectamine 2000 according to the instruction manuals. Two days after transfection, 293T cells were washed with prechilled PBS and scraped with RIPA buffer (Beyotime Biotechnology, China). The lysate from heavy- and light-labeled cells was immunoprecipitated by using anti-Flag magnetic beads (Thermo Fisher Scientific). The beads were washed three times with RIPA buffer and eluted with buffer containing 200 mM Tris-HCl (pH 7.9), 0.2 mg/ml Flag peptide (Sigma), and protease inhibitor cocktail. The eluted proteins were denatured and separated with 10% SDS‒PAGE. A few elutes were used for silver staining analysis. The majority of the eluates were mixed at a 1:1 ratio and stained with Coomassie Brilliant Blue. The combined mixture gel lanes were excised, and in-gel trypsin digestion was performed. The peptides were collected, and LC‒MS/MS was performed on an LTQ-Orbitrap XL mass spectrometer (Thermo Fisher Scientific). The resulting raw data files were analyzed with MaxQuant software against the SWISSPROT database for humans. The peptide/protein ratios were calculated using the SILAC function module.

DDX20 knockout stable cell clones. For genetic knockout of DDX20, two sgRNAs localized in DDX20 exon 1 were independently cloned into BbsI-digested px459 v2.0 (Addgene #62988). Eight micrograms of each sgRNA recombinant vector was then transiently transfected into 1.5 x 10⁶ MDA-MB-231 cells seeded in 10-cm dishes using reduced serum media (Opti-MEM) and Lipofectamine 2000 (Life Technologies). At 24 h posttransfection, 1.5 µg/mL puromycin was added to the cells for two days of selection, and parental cells without plasmids served as the negative control. A portion of these puromycin-resistant enriched cells was harvested for DDX20 detection by Western blotting, and the remaining cells were counted with a Vi-Cell XR Cell Viability Analyzer (Beckman Coulter) and plated at dilutions of 1.5 cells/well in 96-well plates to facilitate the isolation of single-cell clones. After 2–3 weeks, clones were expanded, and DDX20 knockout-positive clone cells were picked and validated by Western blotting.

Generation of MTOR promoter constructs. Genomic DNA isolated from MDA-MB-231 cells was used as a PCR template. A total of 725 bp of the human MTOR promoter [-600, + 107] fragment was amplified by PCR. The fragment was subcloned into a pGL3-basic vector (Promega) at the MluI/XhoI sites. The MTOR promoter reporter with ETV3 binding site mutation was created by directly synthesizing the template by the Synbio-tech company (Suzhou, China) and then cloning it into the pGL3-basic vector.

Splicing reporter. For the splicing reporter plasmid, an intron-containing luciferase reporter, CMV-LUC2CP/ARE (Addgene plasmid #62857), was generated by inserting a 133-nucleotide chimeric intron (pCI-neo, Promega).

Dual luciferase assay. Luciferase-containing reporter plasmids were cotransfected into MDA-MB-231 or HCC-70 cells with the Renilla luciferase vector pRL-SV40 using Lipofectamine 2000. The activity of the reporter was measured using the Dual-Luciferase Reporter Assay System (Promega) after 24 h of transfection and normalized using Renilla luciferase activity in a Molecular Devices Microplate Reader (Spectra Max M5). The results were obtained from three independent experiments.

Chromatin immunoprecipitation assays. Chromatin immunoprecipitation (ChIP)-qPCR assays were performed using an EZ-Magna ChIP A/G kit (Millipore) according to the standard protocol using rabbit polyclonal anti-ETV3 and anti-DDX20. The precipitated DNA samples were analyzed by a quantitative PCR assay, and the data are shown as percentages of the input DNA. Anti-rabbit IgG (Santa Cruz Biotechnology) was used as a control.

In vivo studies. For orthotopic primary tumor formation, shSNRPE #1 tet-on MDA-MB-231 cells were washed with PBS and resuspended in PBS buffer with an equal amount of Matrigel (BD Biosciences, San Diego, CA, United States). Six- to eight-week-old female BALB/c nude mice were injected into the 4th mammary fat pad with 100 µl suspension containing 5.0 × 10⁶ cancer cells. Tumor volume was monitored at 2 weeks after implantation. The volumes were calculated with the formula (mm³) = width × width × length × 0.5, and animals were sacrificed if tumor sizes reached 2000 mm³ for animal ethics. For tail vein injection, luciferase-expressing LM2-4175 cells with tet-on shSNRPE #1 (1.0 × 10⁵ cells in 100 µL PBS) were injected per mouse. The formation of lung metastasis was monitored by bioluminescence imaging using the NightOwl LB981 imaging system. On day 14, to induce SNRPE silencing in vivo, 2 g/L doxycycline was added to the drinking water of mice after grouping according to the normalized photon flux. All animal studies were conducted according to the guidelines of the ethics committee.

RNA-seq. Total RNA was extracted using the RNeasy Mini Kit (QIAGEN) following the instructions. One microgram of total RNA was treated with VAHTS mRNA Capture Beads (Vazyme, Nanjing, China) for the enrichment of polyA + RNA. RNA-seq libraries were performed using a VAHTS mRNA-seq v2 Library Prep Kit for Illumina (Vazyme, Nanjing, China) according to the manufacturer’s instructions. Briefly, enriched RNA samples were fragmented and reverse copied into first- and second-strand cDNA with random hexamer primers. After treatment with end repair and dA-tailing, the dsDNA was ligated with an adapter. The dsDNA was purified and subjected to 12 cycles of PCR amplification, and the libraries were constructed and sequenced on an Illumina sequencing platform. The RNA-seq raw data of sequencing reads were aligned using the spliced read aligner HISAT2 and then mapped to the human reference genome. For alternative splicing (AS) analysis, human genome references of AS events were downloaded from the MISO website and operated according to MISO manuals. The output data were filtered on the basis of both PSI (percent spliced in) differences and PSI distribution plots followed a sashimi plot.

Statistical analysis. Student’s t test was used to compare continuous variables. We used a chi-square test for categorical variables and a chi-square or Fisher’s exact test for contingency tables. The Kaplan‒Meier method was used to compare the clinical outcome between the high-risk and low-risk groups, and the P value was calculated by the log-rank test. Adjusted hazard ratios (HRs) with 95% confidence intervals (CIs) were calculated using Cox proportional hazards modeling. All other statistical tests were two-sided. Statistical significance is shown with NS (not significant) (p > 0.05), *p value ≤ 0.05, **p value ≤ 0.01, *** p value ≤ 0.001. The statistical analyses were performed using SPSS 22.0 (SPSS Inc., Chicago, IL, USA) or R 3.5.0 (R Foundation for Statistical Computing, Vienna, Austria) software programs.

Identification of SNRPE as a candidate for core spliceosome vulnerability in TNBC

To explore why MYC-driven TNBC was extremely sensitive to spliceosome-targeted therapy, we focused specifically on the seven core spliceosomal Sm genes. We first analyzed CNV data to characterize seven Sm genes in TNBC. Among these 465 patients, a total of 401 had CNV data. We found that there is a substantial divergence of Sm gene frequency in TNBC. SNRPE was the most highly amplified Sm gene (250/401, 62%, Fig. 1A). We divided the patients into two groups (Amp/gain and Non-Amp/gain) according to the copy number variation of each Sm gene. Then, we performed univariate Cox proportional hazards regression analysis and found that only copy number amplification or gain of SNRPE was associated with poor prognosis among the seven Sm genes (Fig. 1B). The log rank analysis showed that the prognosis of the SNRPE Amp/gain group was significantly worse than that of the other group (P < 0.05, Fig. 1C). In addition, we analyzed the RNA-seq data of 88 paired samples of cancer and adjacent tissues in our TNBC multiomics database. The results showed that the expression of SNRPE in TNBC was significantly higher than that in adjacent tissues (Fig. 1D). Moreover, according to the Fudan "four subtypes" of TNBC proposed by our center [4], we discovered that the expression of SNRPE in the BLIS subtype with poor prognosis was significantly higher than that in the other three subtypes (Fig. 1E). Furthermore, log rank survival analysis showed that in TNBC patients, the prognosis of the SNRPE high expression group was significantly worse than that of the SNRPE low expression group (P < 0.001, Fig. 1F, S2). To further confirm these results, we included 309 TNBC patients in the METABRIC database as a validation cohort to support our results (Fig. S1A-C). Similarly, based on the expression profile data of TNBC in TCGA database, we found that the expression of SNRPE in TNBC was significantly higher than that in adjacent cancer (Fig. S1D), and by TCGA pancancer analysis, interestingly, SNPRE was also highly expressed in all solid tumors (Fig. 1G), as well as other Sm genes (Fig. S1F), and this may be a common feature of solid tumors. In addition, we analyzed SNRPE expression and recurrence-free survival of TNBC in a public database and found that high SNRPE expression was a poor prognostic factor (P < 0.05, Fig. S1E). In summary, in the validation set and the training set, SNRPE was significantly amplified and highly expressed in TNBC patients with poor prognosis.

Characterization of SNRPE in TNBC

Previously, the SNRPE gene was reported to contain pseudogenes in the human genome [17]. We searched at least 10 pseudogenes, two of which (SNRPEP2 and SNPRPEP4) may be active (Fig. S3). Using cDNAs from two breast cancer cell lines as templates, we designed a pair of compatible PCR primers to amplify the transcripts followed by high-throughput sequencing (Fig. S4A). All pseudogenes were not transcriptionally active; instead, we identified two novel unreported noncoding transcripts of SNPRE with extended frame-shift Exon 2, T1_P1 and T1_P2 (Fig. S4B, S4C). Our observation of the MDA-MB-231 cell line revealed that the RNA-seq peak of SNPRE did not correspond to the annotated NM_003094 and was a much shorter 3’UTR sequence (Fig. S4D), and this result was also confirmed in our clinical TNBC tissues (Fig. S4E). The 3' RACE results were perfectly consistent with the RNA-seq data, while the PAS software prediction also confirmed the presence of better new PAS sites (Fig. S4F). Microarray annotation of the SNRPE probe set should be 203316_s_at rather than 231112_at in TNBC (Fig. S1E, S4D). A shorter 3’UTR may avoid negative regulation to stabilize SNPRE mRNA, such as microRNA modulation. In conclusion, in TNBC, pseudogene-encoded SNPRE transcripts do not exist, and the function of SNPRE is essentially dependent on protein function.

SNPRE expression silencing leads to tumor-selective growth arrest both in vitro and in vivo

To explore the role of SNRPE in TNBC, we first detected SNRPE expression in two immortalized normal breast epithelial cell lines and different subtypes of breast cancer cell lines and found that SNRPE was highly expressed in breast cancer cells relative to HMEC normal breast epithelial cells, both at the mRNA and protein levels, with higher expression in TNBC cell lines (Fig. S5A, 2A). We selected HCC70 and MDA-MB-231 for the TNBC study, and MCF10A, a breast epithelial cell line with slightly higher SNPRE expression, was also included in the study. First, we selected two effective shRNA interference target sequences to construct SNRPE-silenced stable cell lines (Fig. 2B). SNRPE knockdown significantly inhibited the proliferation of TNBC cell lines (Figure 2D, 2F), and overexpression of SNRPE enhanced growth (Fig. S5B, S5C). However, in breast epithelial MCF10A cells, SNRPE knockdown had little effect on cell proliferation (Figure 2H). Meanwhile, SNRPE knockdown significantly reduced the migration capacity in TNBC cell lines (Figure 2E, 2G), and SNPRE upregulation also increased migration (Fig. S5D). Given the sharp inhibition of cell proliferation following SNRPE knockdown, to obtain sufficient tumor cells for the in vivo assay in mice, we used the tetracycline-inducible gene expression system (tet-on) to induce SNRPE knockdown (Figure 2C). We established a mammary fat pad injection model in female BALB/c nude mice. Mice fed doxycycline had a significant reduction in tumor size (Figure 2I). In addition, we induced SNRPE downregulation in LM2-4175 cells, MDA-MB-231-derived clone cells with high lung metastatic potential (Fig. S5E). After tail vein injection, we found that the growth rate of lung metastases in the SNRPE-induced knockdown group was significantly lower than that in the control group (Figure S5F). In conclusion, SNRPE knockdown significantly inhibited the proliferation and migration of TNBC cell lines in vitro and in vivo.

Establishment of the interacting protein DDX20 as a key molecule for vulnerability to spliceosome inhibition in TNBC

To uncover why SNPRE downregulation is tumor-selective killed, among the SNPRE interacting proteins, we hope to identify a candidate protein that must fulfill a condition: differential expression in tumor and normal tissues. We used stable isotope labeling of amino acids in cell culture (SILAC) combined with quantitative proteomics techniques to search for molecules interacting with SNRPE in HEK-293T cell lines. We successfully constructed flag-tagged SNRPE overexpression plasmids (Fig. 3A) and identified approximately 60 interacting proteins by mass spectrometry (more than 1.5-fold difference, more than 2 unique peptides, Table S2). Among these interacting proteins, in addition to the core spliceosome components containing other Sm proteins, another SMN protein complex (SMN1, GEMIN2, DDX20, GEMIN4) involved in the maturation of spliceosome processing is highlighted (Fig. 3B). DDX20 was previously reported to be highly expressed to promote multiorgan metastasis in TNBC [12]. The expression pattern of DDX20 is similar to that of the MYC gene in breast cancer subtypes (Fig. S6A) because DDX20 is a downstream gene of MYC [18]. In addition, DDX20 is highly expressed in the testis and may be a cancer-testis gene (Fig. S6B) [19], so we first attempted to choose DDX20 as a downstream molecule to explore the oncogenic mechanism of SNPRE. In MDA-MB-231 and HCC70 cell lines, the interaction between SNRPE and DDX20 was confirmed by coimmunoprecipitation (Co-IP) (Fig. 3C, 3D). We transiently transfected breast cancer cell lines to obtain effective siRNAs (Fig. 3E), and knockdown of DDX20 had no effect on tumor growth in TNBC (Fig. 3F). Hence, we considered DDX20 knockout by the CRISPR technique, and we successfully isolated two clones from sgDDX20#2 (Fig. S6D). Similarly, DDX20 knockout also had no significant effect on tumor growth (Figure S6D). Taken together, DDX20, as a potentially synergistic molecule involved in core spliceosome vulnerability, may be in a previously unknown way.

The DDX20-SNRPE axis is involved in tumor-selective killing independent of RNA splicing

Both DDX20 and SNPRE participate in spliceosome SnRNP biosynthesis, with SNRPE altering the growth phenotype but not DDX20, so we questioned whether SNPRE regulates alternative splicing of some reported cancer-related genes. As a result, we found a few genes with alternative splicing changes after SNREP KD and did not identify oncogenes or tumor-suppress genes in the literature (Fig. 4A, Table S3). Here, we used effective transient siRNA instead of stable shRNA to perform KD functions to avoid the indirect splicing events caused by downstream genes (Fig. 4B). We consider that SNPRE dysregulation may affect tumor phenotypes through constitutive splicing rather than alternative splicing. Then, we utilized a constitutive splicing reporter vector (Fig. 4C). When either SNRPE or DDX20 was downregulated, constitutive splicing activity was significantly reduced in TNBC (Fig. 4D). A double downregulated DDX20-KO/SNRPE-KD cell line was obtained by infection with SNPRE-interfering lentivirus in one of the DDX20 knockout clones (Fig. 4E). We found that in double downregulated cells, their splicing capacity was further reduced (Fig. 4F), but surprisingly, the cell growth of TNBC cells was nearly unaffected (Fig. 4G). In summary, we propose a hypothesis that SNRPE expression silencing leading to TNBC cell growth arrest is dependent on DDX20 expression and independent of RNA splicing.

The SNPRE-DDX20 axis regulates tumor proliferation via MTOR signaling

To explore this mechanism, we first performed gene set enrichment analysis (GSEA) of RNA-seq data using SNRPE KD and DDX20 KO knockout cells as well as double downregulated (DD) cells. Only SNPRE knockdown caused growth arrest because the PI3K/AKT/mTOR (PAM) pathway was blocked (Fig. 5A), whereas in DDX20 KO or DD cells, the PAM pathway was still in an activated state (Fig. 5B). These findings suggest that SNRPE promotes TNBC via the PAM pathway and is dependent on DDX20, and DDX20 is indeed a key molecule that confers vulnerability to spliceosome-targeted therapy. SNPRE has previously been reported in luminal breast cancer, and mRNA gene microarray data showed that MTOR was one of the most significantly differentially expressed genes after SNPRE downregulation [20], which we have also observed here in TNBC cells. MTOR is the catalytic unit in the mTORC1 and mTORC2 complexes involved in the cell growth pathway [21], and MTOR silencing caused tumor growth arrest more significantly because MTORC1 and MTORC2 were simultaneously damaged [22]. In SNRPE KD cells, the protein level of MTOR was significantly reduced, along with its major downstream effector p-4EBP1 (Fig. 5C). When DDX20 was knocked out, downregulation of SNRPE was no longer able to reduce MTOR expression (Fig. 5D), and a slight increase in MTOR protein resulted in maintenance of MTOR pathway activation to promote tumor growth (Fig. 5E). In conclusion, SNPRE activates the MTOR pathway by directly positively regulating MTOR transcription in a DDX20-dependent manner.

The SNPRE-DDX20 axis affects MTOR transcription directly via the transcription factor ETV3

In addition to the splicing process, DDX20 has also been reported to act as a transcriptional cofactor [23]. Therefore, we first considered whether the MTOR gene is transcriptionally regulated by DDX20. Using ChIP data from the Cistrome database (http://cistrome.org/db/#/), we found that DDX20 has a strong unique binding signal in the promoter of MTOR (Fig. 6A). DDX20 acts as a cofactor due to the loss of the DNA-binding domain. We performed JASPAR transcription factor prediction with the 217 bp DNA sequence of the MTOR promoter co-occupied with DDX20 to identify potential transcription factors (Table S1, S4). The ETS family of transcription factors is highlighted, and ETV3 (also known as METS and PE1) has previously been reported to interact with DDX20 [23]. Another study also reported that there is a correlation between MYC and ETV3 transcript levels in breast cancer. ETV3 was also the most abundantly expressed ETV protein in breast cancer and was highly expressed in TNBC (Fig. S7A). Therefore, we first validated whether ETV3 is a downstream transcriptional molecule of the DDX20-SNRPE axis that regulates MTOR transcription. To test this speculation, we carried out and confirmed that endogenous anti-ETV3 coimmunoprecipitated with endogenous DDX20 from TNBC cell lines of whole cell extracts (Fig. 6B, Fig. S7B). ChIP assays showed that both DDX20 and ETV3 can bind to the promoter of MTOR (Fig. 6C, S7C). In DDX20 KO cells, ETV3 expression was not affected (Fig. 6D), but the binding affinity of ETV3 in the MTOR promoter region was significantly reduced (Fig. 6E). To detect MTOR promoter activity, we constructed an MTOR promoter reporter as well as a reporter with a mutated ETV3 binding site (Fig. 6F). DDX20 knockout resulted in a decrease in ETV3 binding affinity in the MTOR promoter but a significant increase in MTOR transcriptional activity (Fig. 6G), thus confirming that DDX20-ETV3 regulation of MTOR is a negative regulation of transcriptional repression. Mutation of the ETV3 binding site of the MTOR promoter greatly enhanced MTOR promoter activity, regardless of DDX20 expression. From the right side of Figure 6A, we clearly see the presence of both an activated cis-element (H3K27ac) and a repressed cis-element (DDX20) in distinctly separate regions in the MTOR promoter. Therefore, a transcriptional repressor cis-element in the MTOR promoter can be occupied with ETV3-DDX20, which subsequently represses MTOR expression.

Core spliceosomal SNRPE antagonizes the negative modulation of MTOR signaling in TNBC

The transcriptional repressor ETV3 together with its corepressor DDX20 can suppress the expression activity of MTOR. How does core spliceosomal SNRPE deregulate transcriptional repression in TNBC cells? Previous studies confirmed that the transcriptional repressive activity of DDX20 is at the C-terminus and that both ETV3 and SNPRE can bind to the C-terminus of DDX20 [24], where the interaction of SNRPE with DDX20 is indirect [25]. The C-terminus of the DDX20 3-D structure remains unresolved and cannot be predicted by homology matching. (Fig. S8). Meanwhile, DDX20 mutations in TCGA data were enriched on the C-terminus of DDX20 (Table S5), implying that the C-terminus may be important in tumorigenesis. Hence, we hypothesize that in TNBC cells, ETV3 interacts with the C-terminal region of DDX20 but is antagonized by SNPRE-containing components, resulting in ETV3 without DDX20 accompaniment and making its transcriptional repressive activity insufficient so that MTOR can be smoothly expressed. When SNRPE was silenced, we confirmed that MTOR promoter activity was significantly decreased in TNBC (Fig. 7A, 7B), and DDX20 expression was nearly unaffected, but the interaction between DDX20 and ETV3 was significantly increased (Fig. 7C, 7D). Additionally, we also found a significant increase in the binding affinity of ETV3 to the MTOR promoter (Fig. 7E, 7F). Thus, SNRPE antagonizes the transcriptional repression of MTOR by attenuating two interacting groups: DDX20-ETV3 and ETV3-DNA. In addition to the canonical pathway involved in spliceosome SnRNP biosynthesis, SNRPE-containing components antagonize the negative regulation of the MTOR pathway by ETV3/DDX20 through a noncanonical splicing-independent pathway (Fig. S9).

In MYC-driven cancers (greater than 30% of cancers), such as TNBC, spliceosome-targeted therapy is sensitive and has become a therapeutic hotspot [4, 26]. Here, we used a core spliceosomal gene, SNRPE, as our study and found that one of its interacting partners, DDX20, is a vital molecule leading to vulnerability to spliceosome inhibition. As shown in Fig. 7G, in MYC-driven cancers, both SNRPE and DDX20 are highly expressed by MYC regulation [7, 18, 27], and DDX20, as a transcriptional corepressor of ETV3, is antagonized by the abundantly expressed SNRPE protein, thus failing to exercise transcriptional repression and allowing proliferation. After SNRPE downregulation, DDX20 is released to bind with ETV3, while the affinity of ETV3 anchoring to the MTOR promoter is also increased, thereby substantially reducing MTOR transcription and causing growth arrest. When both DDX20 and SNRPE were downregulated, RNA splicing activity decreased dramatically, and surprisingly, cancer cells still survived due to MTOR expression recovery, a state similar to normal cells with low expression of both DDX20 and SNRPE. With low expression of DDX20 in normal tissues, downregulation of SNRPE by RNA interference could not affect growth either. The development of novel drugs that interrupt the interaction between DDX20 and SNRPE may inhibit spliceosome-mediated RNA splicing to induce intron retention to trigger immunotherapy and turn off the MTOR pathway to directly inhibit tumor proliferation.

SNRPE, as a small protein of only 10.8 kD, belongs to the family of Sm and Sm-like proteins, which are conserved in archaebacteria and eukaryotes [28]. SNPRE has been previously reported to be involved in the tumorigenesis of several cancers, such as luminal breast cancer, liver cancer, prostate cancer, and lung cancer [20, 29–31]. For instance, in prostate cancer, SNPRE was the most significantly upregulated gene expressed in prostate cancer as well as the AR gene [32], where AR expression was reported to be transcriptionally regulated by SNPRE, and this regulation also appears to be independent of the splicing process [29]. Conversely, Z. Li et al. reported that SNRPE is one of the E2F1-regulated genes, and SNRPE can be involved in p53-independent pro-apoptotic functions in tumors. Downregulation of endogenous SNPRE can promote tumor proliferation, and presumably, there is a splicing-independent mechanism. However, the conclusion that SNRPE functions as a tumor suppressor gene rather than an oncogene is debatable [33, 34]. Other studies have also shown the presence of other nonsplicing physiological activities in SNPRE. The processing of SNRPE by the SMN complex takes place in the cytosol, whereas the involvement of SMN complex components and SNPRE in transcriptional regulation has been documented in the nucleus [35]. In Xenopus laevis, Sm proteins are also present in nuage and mitochondrial cement involved in splicing-independent functions [36]. In nematodes, a novel function of SNRPE was found to be closely related to embryogenesis, where the function was defined as transcriptional silencing [37, 38]. Yeast one-hybrid assays demonstrated that SNPRE could be transferred to the promoter region of Hsp70, an intronless gene, to promote the transcriptional involvement of another protein, Hsp105beta [39]. In summary, SNPRE participates in splicing-independent functions, but the mechanism is still elusive and requires subsequent in-depth mechanistic studies.

DDX20 is a multifunctional protein that engages in various cellular processes, whether oncogenic or tumor-suppressive, in particular cancer functions through the common NF-κB pathway. In hepatocellular carcinoma, DDX20 is a tumor suppressor gene that is expressed at low levels in cancer cells and inhibits NF-κB activity by preferentially regulating NF-κB-suppressing miRNA-140 [40, 41]. In contrast, DDX20 is highly expressed in breast cancer and functions as a metastasis-promoting gene that activates NF-κB activity by interacting with TAK1 [12], while our data here suggest that DDX20 is not essential for breast cancer growth. In prostate cancer, DDX20 promotes proliferation and metastatic potential through the NF-κB pathway [42].

Several core spliceosomal components harboring additional spliceosome-independent or splicing-independent functions have also been reported, and generally, through their RBP domains, nascent transcripts, such as R-loops or specific structured RNA elements, are bound [43, 44]. Among ~ 200 known splicing-related components, in addition to DDX20 and SNPRE, whether other components, such as common interacting partners of DDX20 and SNPRE (i.e., SMN, GEMIN2, GEMIN4, other Sm) participate in transcriptional repression is still unknown, and our subsequent work will focus on this detailed mechanism. Remarkably, the C-terminus of DDX20 is required for both ETV3 and SNRPE, and fine resolution or mapping of the interaction sites between them could lead to a better understanding of the mechanisms and drug development.

Our results reveal that in MYC-driven tumors, MYC drives the expression of a range of downstream molecules, such as numerous spliceosome-related genes (e.g., SNRPE and DDX20), and that there is a complex regulatory relationship between these downstream molecules. Direct targeting of MYC to downregulate the expression of all downstream molecules (e.g., cells with dual downregulation of DDX20 and SNRPE) may not completely inhibit cancer growth. DDX20 is an adaptor in the crosstalk between RNA splicing activity (SnRNP components such as SNRPE) and the growth pathway (MTOR expression). Theoretically, cancer patients with high DDX20 expression are more suitable for treatment with spliceosomal drugs, which both directly inhibit tumor growth and increase antitumor immunity. Targeting DDX20 may serve as a therapeutic target for MYC gene-driven tumors.

TNBC: triple-negative breast cancer; STT: spliceosome-targeted therapy; SnRNP: small nuclear ribonucleoprotein; RACE: rapid amplification of cDNA ends; siRNA: small interfering RNA; shRNA: short hairpin RNA; ChIP: chromatin immunoprecipitation; HR: hazard ratio; CI: confidence interval; CNV: copy number variation; RFS: relapse-free survival; GO: Gene Ontology; IHC: immunohistochemistry; PAS: polyadenylation site; UTR: untranslated region; SILAC: stable isotope labeling of amino acids in cell culture; CoIP: coimmunoprecipitation; KD: knockdown; KO: knockout; DD: double downregulation; GSEA: gene set enrichment analysis; PAM: PI3K/AKT/mTOR;

Ethics approval and consent to participate: This study was approved by the Ethics Committee of FUSCC. The animal protocols were approved by the Animal Welfare Committee of Shanghai Medical College at Fudan University.

Consent for publication: The content of this manuscript has not been previously published and is not under consideration for publication elsewhere. All authors approved the final version of the manuscript.

Competing interests: The authors declare no competing financial interests.

Funding: This work was supported by the National Natural Science Foundation of China (92159301, 91959207, 81572583, 81672600, 81302301), the Cooperation Project of Conquering Major Diseases in the Shanghai Municipality Health System (2013ZYJB0302), and the Shanghai Key Laboratory of Breast Cancer (12DZ2260100).

Author contributions: G.-D.X. and X.-E.X. conducted and analyzed most of the experiments. X.-E.X., Z.-M.S., and K.-D.Y. designed the study. X.J. and W.-T.P. conducted the remaining experiments. X.-E.X., and G.-D.X. prepared the manuscript.

Acknowledgments: We thank the Department of Breast Surgery, Fudan University Shanghai Cancer Center, for their excellent data management.

Availability of data and materials

All data generated or analyzed are available from the corresponding author upon reasonable request: X.-E.X. ([email protected]).

Rahman MA, Krainer AR, Abdel-Wahab O: SnapShot: Splicing Alterations in Cancer. Cell 2020, 180(1):208-208 e201.
Yang H, Beutler B, Zhang D: Emerging roles of spliceosome in cancer and immunity. Protein Cell 2021.
Anczuków O, Krainer AR: The spliceosome, a potential Achilles heel of MYC-driven tumors. Genome Med 2015, 7:107.
Jiang YZ, Ma D, Suo C, Shi J, Xue M, Hu X, Xiao Y, Yu KD, Liu YR, Yu Y et al: Genomic and Transcriptomic Landscape of Triple-Negative Breast Cancers: Subtypes and Treatment Strategies. Cancer Cell 2019, 35(3):428-440 e425.
Petrocca F, Altschuler G, Tan SM, Mendillo ML, Yan H, Jerry DJ, Kung AL, Hide W, Ince TA, Lieberman J: A genome-wide siRNA screen identifies proteasome addiction as a vulnerability of basal-like triple-negative breast cancer cells. Cancer Cell 2013, 24(2):182-196.
Marcotte R, Sayad A, Brown KR, Sanchez-Garcia F, Reimand J, Haider M, Virtanen C, Bradner JE, Bader GD, Mills GB et al: Functional Genomic Landscape of Human Breast Cancer Drivers, Vulnerabilities, and Resistance. Cell 2016, 164(1-2):293-309.
Hsu TY, Simon LM, Neill NJ, Marcotte R, Sayad A, Bland CS, Echeverria GV, Sun T, Kurley SJ, Tyagi S et al: The spliceosome is a therapeutic vulnerability in MYC-driven cancer. Nature 2015, 525(7569):384-388.
Blijlevens M, van der Meulen-Muileman IH, de Menezes RX, Smit EF, van Beusechem VW: High-throughput RNAi screening reveals cancer-selective lethal targets in the RNA spliceosome. Oncogene 2019, 38(21):4142-4153.
Bowling EA, Wang JH, Gong F, Wu W, Neill NJ, Kim IS, Tyagi S, Orellana M, Kurley SJ, Dominguez-Vidaña R et al: Spliceosome-targeted therapies trigger an antiviral immune response in triple-negative breast cancer. Cell 2021, 184(2):384-403 e321.
Lu SX, De Neef E, Thomas JD, Sabio E, Rousseau B, Gigoux M, Knorr DA, Greenbaum B, Elhanati Y, Hogg SJ et al: Pharmacologic modulation of RNA splicing enhances anti-tumor immunity. Cell 2021, 184(15):4032-4047 e4031.
Jung H, Lee D, Lee J, Park D, Kim YJ, Park WY, Hong D, Park PJ, Lee E: Intron retention is a widespread mechanism of tumor-suppressor inactivation. Nat Genet 2015, 47(11):1242-1248.
Shin EM, Hay HS, Lee MH, Goh JN, Tan TZ, Sen YP, Lim SW, Yousef EM, Ong HT, Thike AA et al: DEAD-box helicase DP103 defines metastatic potential of human breast cancers. J Clin Invest 2014, 124(9):3807-3824.
Bizen N, Bepari AK, Zhou L, Abe M, Sakimura K, Ono K, Takebayashi H: Ddx20, an Olig2 binding factor, governs the survival of neural and oligodendrocyte progenitor cells via proper Mdm2 splicing and p53 suppression. Cell Death Differ 2022, 29(5):1028-1041.
Wang Q, Ye Y, Lin R, Weng S, Cai F, Zou M, Niu H, Ge L, Lin Y: Analysis of the expression, function, prognosis and co-expression genes of DDX20 in gastric cancer. Comput Struct Biotechnol J 2020, 18:2453-2462.
Huang Y, Wang C, Li K, Ye Y, Shen A, Guo L, Chen P, Meng C, Wang Q, Yang X et al: Death-associated protein kinase 1 suppresses hepatocellular carcinoma cell migration and invasion by upregulation of DEAD-box helicase 20. Cancer Sci 2020, 111(8):2803-2813.
Chen W, Zhou P, Li X: High expression of DDX20 enhances the proliferation and metastatic potential of prostate cancer cells through the NF-κB pathway. Int J Mol Med 2016, 37(6):1551-1557.
Stanford DR, Holicky EL, Perry CA, Rehder K, Jr., Harvey SE, Rohleder AM, Wieben ED: The snRNP E protein multigene family contains five pseudogenes with common mutations. DNA Seq 1991, 1(5):357-363.
Koh CM, Bezzi M, Low DH, Ang WX, Teo SX, Gay FP, Al-Haddawi M, Tan SY, Osato M, Sabò A et al: MYC regulates the core pre-mRNA splicing machinery as an essential step in lymphomagenesis. Nature 2015, 523(7558):96-100.
Mac Keon S, Bentivegna S, Levy EM, Marks MS, Mantegazza AR, Wainstok R, Mordoh J: Syngeneic B16-F1 cells are more efficient than allogeneic Cloudman cells as antigen source in DC-based vaccination in the B16-F1 murine melanoma model. Vaccine 2019, 37(35):4947-4955.
Quidville V, Alsafadi S, Goubar A, Commo F, Scott V, Pioche-Durieu C, Girault I, Baconnais S, Le Cam E, Lazar V et al: Targeting the deregulated spliceosome core machinery in cancer cells triggers mTOR blockade and autophagy. Cancer Res 2013, 73(7):2247-2258.
Hung CM, Garcia-Haro L, Sparks CA, Guertin DA: mTOR-dependent cell survival mechanisms. Cold Spring Harb Perspect Biol 2012, 4(12).
Montero JC, Esparís-Ogando A, Re-Louhau MF, Seoane S, Abad M, Calero R, Ocaña A, Pandiella A: Active kinase profiling, genetic and pharmacological data define mTOR as an important common target in triple-negative breast cancer. Oncogene 2014, 33(2):148-156.
Klappacher GW, Lunyak VV, Sykes DB, Sawka-Verhelle D, Sage J, Brard G, Ngo SD, Gangadharan D, Jacks T, Kamps MP et al: An induced Ets repressor complex regulates growth arrest during terminal macrophage differentiation. Cell 2002, 109(2):169-180.
Curmi F, Cauchi RJ: The multiple lives of DEAD-box RNA helicase DP103/DDX20/Gemin3. Biochem Soc Trans 2018, 46(2):329-341.
Charroux B, Pellizzoni L, Perkinson RA, Shevchenko A, Mann M, Dreyfuss G: Gemin3: A novel DEAD box protein that interacts with SMN, the spinal muscular atrophy gene product, and is a component of gems. J Cell Biol 1999, 147(6):1181-1194.
Dang CV, Kim J-w, Gao P, Yustein J: The interplay between MYC and HIF in cancer. Nature Reviews Cancer 2008, 8(1):51-56.
Das S, Anczuków O, Akerman M, Krainer AR: Oncogenic splicing factor SRSF1 is a critical transcriptional target of MYC. Cell Rep 2012, 1(2):110-117.
Mayes AE, Verdone L, Legrain P, Beggs JD: Characterization of Sm-like proteins in yeast and their association with U6 snRNA. EMBO J 1999, 18(15):4321-4331.
Anchi T, Tamura K, Furihata M, Satake H, Sakoda H, Kawada C, Kamei M, Shimamoto T, Fukuhara H, Fukata S et al: SNRPE is involved in cell proliferation and progression of high-grade prostate cancer through the regulation of androgen receptor expression. Oncol Lett 2012, 3(2):264-268.
Jia D, Wei L, Guo W, Zha R, Bao M, Chen Z, Zhao Y, Ge C, Zhao F, Chen T et al: Genome-wide copy number analyses identified novel cancer genes in hepatocellular carcinoma. Hepatology 2011, 54(4):1227-1236.
Liu G, Li F, Chen M, Luo Y, Dai Y, Hou P: SNRPD1/E/F/G Serve as Potential Prognostic Biomarkers in Lung Adenocarcinoma. Front Genet 2022, 13:813285.
Tamura K, Furihata M, Tsunoda T, Ashida S, Takata R, Obara W, Yoshioka H, Daigo Y, Nasu Y, Kumon H et al: Molecular Features of Hormone-Refractory Prostate Cancer Cells by Genome-Wide Gene Expression Profiles. Cancer Research 2007, 67(11):5117-5125.
Li Z, Stanelle J, Leurs C, Hanenberg H, Pützer BM: Selection of novel mediators of E2F1-induced apoptosis through retroviral expression of an antisense cDNA library. Nucleic Acids Res 2005, 33(9):2813-2821.
Li Z, Pützer BM: Spliceosomal protein E regulates neoplastic cell growth by modulating expression of cyclin E/CDK2 and G2/M checkpoint proteins. J Cell Mol Med 2008, 12(6A):2427-2438.
Meister G, Bühler D, Laggerbauer B, Zobawa M, Lottspeich F, Fischer U: Characterization of a nuclear 20S complex containing the survival of motor neurons (SMN) protein and a specific subset of spliceosomal Sm proteins. Human Molecular Genetics 2000, 9(13):1977-1986.
Bilinski SM, Jaglarz MK, Szymanska B, Etkin LD, Kloc M: Sm proteins, the constituents of the spliceosome, are components of nuage and mitochondrial cement in Xenopus oocytes. Experimental Cell Research 2004, 299(1):171-178.
Barbee SA, Lublin AL, Evans TC: A Novel Function for the Sm Proteins in Germ Granule Localization during C. elegans Embryogenesis. Current Biology 2002, 12(17):1502-1506.
Barbee SA, Evans TC: The Sm proteins regulate germ cell specification during early C. elegans embryogenesis. Developmental Biology 2006, 291(1):132-143.
Saito Y, Nakagawa T, Kakihana A, Nakamura Y, Nabika T, Kasai M, Takamori M, Yamagishi N, Kuga T, Hatayama T et al: Yeast Two-Hybrid and One-Hybrid Screenings Identify Regulators of hsp70 Gene Expression. J Cell Biochem 2016, 117(9):2109-2117.
Takata A, Otsuka M, Yoshikawa T, Kishikawa T, Kudo Y, Goto T, Yoshida H, Koike K: A miRNA machinery component DDX20 controls NF-κB via microRNA-140 function. Biochem Biophys Res Commun 2012, 420(3):564-569.
Takata A, Otsuka M, Yoshikawa T, Kishikawa T, Hikiba Y, Obi S, Goto T, Kang YJ, Maeda S, Yoshida H et al: MicroRNA-140 acts as a liver tumor suppressor by controlling NF-κB activity by directly targeting DNA methyltransferase 1 (Dnmt1) expression. Hepatology 2013, 57(1):162-170.
Chen W, Zhou P, Li X: High expression of DDX20 enhances the proliferation and metastatic potential of prostate cancer cells through the NF-κB pathway. Int J Mol Med 2016, 37(6):1551-1557.
Chen L, Chen JY, Huang YJ, Gu Y, Qiu J, Qian H, Shao C, Zhang X, Hu J, Li H et al: The Augmented R-Loop Is a Unifying Mechanism for Myelodysplastic Syndromes Induced by High-Risk Splicing Factor Mutations. Mol Cell 2018, 69(3):412-425 e416.
Fish L, Khoroshkin M, Navickas A, Garcia K, Culbertson B, Hänisch B, Zhang S, Nguyen HCB, Soto LM, Dermit M et al: A prometastatic splicing program regulated by SNRPA1 interactions with structured RNA elements. Science 2021, 372(6543).

No competing interests reported.

Fig.S1.tif
Additional file 1: Supplementary Figure 1. SNRPE was highly expressed in public TNBC cohorts and associated with poor prognosis. (A) The copy number variations of Sm genes in the METABRIC database (n=309; amplification, gain, neutral, loss, and deletion were specified as GISTIC +2, +1, 0, -1, and -2). (B) Association of the Sm gene copy number variations with RFS according to Cox regression analyses in the METABRIC database. (C) Kaplan‒Meier survival curves of RFS for TNBC patients with or without SNPRE amplification or gain in the METABRIC database. (D) The gene expression (log2(FPKM + 1)) of SNRPE in TNBC and normal samples in TCGA database. (E) Kaplan‒Meiersurvival curves of TNBC patients with high or low levels of SNRPE at http://www.kmplot.com/analysis/. (F) TCGA RNA-seq data analysis of the expression of other Sm genes in solid tumors in carcinoma and para-cancer obtained from the GEPIA database.
Fig.S3.tif
Additional file 3: Supplementary Figure 3. Distribution of 10 SNRPE pseudogenes in the human genome. A BLAST search of the human genome with SNRPE cDNA NM_003094. This revealed a total of 10 pseudogene loci. Two of the ten SNPRE pseudogenes, SNRPEP2 and SNRPEP4, with both microarray probe set annotation and RNA-seq mapping, are highlighted in red font.
Fig.S4.tif
Additional file 4: Supplementary Figure 4. Characterization of SNRPE transcripts in TNBC (A) A multiple sequence alignment of parental SNRPE and its two pseudogenes was performed via Clustal Omega software online with default parameters. The area in red is the common replicon amplified by a pair of compatible PCR primers. (B) PCR products amplified from a pair of compatible primers are shown by pie chart distribution. (C) The detailed information of enriched PCR products is shown by comparison with the SNPRE gene locus. (D) The SNRPE probe array contains 203316_s_at probe sets and 231112_at probe sets. The RNA-seq of MDA-MB-231 (self-sequencing, data not shown here) is placed in the SNPRE gene locus. (E) Three random RNA-seq peaks of TNBC tissues (data from SRP157974) in the SNPRE gene locus. (F) The new PAS site of SNPRE was predicted by DNAFSMiner software online and experimentally sequenced by 3’ RACE.
Fig.S5.tif
Additional file 5: Supplementary Figure 5. Dysregulation of SNRPE contributes to growth and metastasis in TNBC (A) RT‒qPCR analysis of SNRPE expression in two human mammary epithelial cell lines (HMEC and MCF-10A) and different breast cancer cell lines. (B, C) Cell growth assays of SNRPE overexpression in MDA-MB-231 and HCC70 cells. The cells were placed in the IncuCyte Live Cell Analysis System after 24 h of pcDNA3.1-flag-SNPRE transfection. (D) Cell migration assays of SNRPE upregulation in MDA-MB-231 cells. After 48 h ofpcDNA3.1-flag-SNPRE transfection, the cells were added to the upper compartment of the Boyden chamber for 8 h followed by cell counting. (E) Measurement of SNRPE by western blot in LM2-4175 cells transfected with doxycycline (dox)-inducible shSNRPE-1 tet-on system at the indicated dox concentration. (F) Lung metastasis of xenografts in BALB/c nude mice was represented from bioluminescent imaging at day 14 and day 42. Normalized photon flux of lung metastasis of SNRPE knockdown MDA-MB-231-LM2 xenografts was calculated in BALB/c nude mice.
Fig.S6.tif
Additional file 6: Supplementary Figure 6. Identification of DDX20 as a potential molecule participating in spliceosome therapeutic vulnerability. (A) The profiles of DDX20 and MYC expression are similar in breast cancer PAM50 subtypes. The data were extracted from the TCGA mRNA array database of breast cancer. (B) Contribution of DDX20 expression in human normal tissues. The result was created by combining the HPA and GTEx transcriptomics datasets. (C) Two DDX20 knockout single cell clones of MDA-MB-231 cells were confirmed by western blot with DDX20 antibody. Two clones were both integrated with the same sgRNA sequence: 5’-GGCTTCAAATGCCGCCGCCA-3’. (D) For cell proliferation assays, two DDX20 KO MDA-MB-231 clones were placed in the IncuCyte Live Cell Analysis System for 84 h.
Fig.S7.tif
Additional file 7: Supplementary Figure 7. ETV3 cooperated with DDX20 to inhibit MTOR expression. (A) Expression distribution of six ETV family members in breast cancer PAM50 subtypes. The data were obtained and analyzed from TCGA database. (B) Interaction of DDX20 and ETV3 in HCC-70 parental cells. An endogenous anti-ETV3 immunoprecipitation experiment was performed followed by immunoblotting with an anti-DDX20 antibody. (C) ChIP‒qPCR assays were performed to enrich the human MTOR promoter sequence using anti-DDX20, anti-ETV3 and control IgG antibodies in HCC-70 cells.
Fig.S9.tif
Additional file 9: Supplementary Figure 9. A schematic model depicting the canonical and noncanonical pathways involving both SNPRE and DDX20. A schematic model depicting the canonical pathway of snRNP assembly mediated by the SMN complex and pCI and the noncanonical pathway of snRNP and SMN complex components involved in cell proliferation via MTOR transcription regulation. The proteins in the dashed box are just guessed. The red arrow only exists in tumor cells. Literature support for direct interaction between two proteins in contact.
SupplementaryTable1.Primersandothernucleicacidsequences.docx
Supplementary Table 1. Primers and other nucleic acid sequences
SupplementaryTable2.SNRPEinteractingproteinsbymassspectrometry.docx
Supplementary Table 2. SNRPE-interacting proteins by mass spectrometry
SupplementaryTable3.Alternativesplicingtoprankevents.docx
Supplementary Table 3. Alternative splicing top rank events
SupplementaryTable4.TranscriptionfactorpredictionscooccupiedwithDDX20attheMTORpromotercore217bpregion.docx
Supplementary Table 4. Transcription factor predictionsco-occupied with DDX20 at the MTOR promoter core 217 bp region
SupplementaryTable5.MutationoftheDDX20Cterminusacrosscancers.docx
Supplementary Table 5. Mutation of the DDX20 C-terminus across cancers
Uncroppedblotsforwesternblotpanels.docx
Supplementary file. Uncropped blots for western blot panels.

Download PDF

Version 1

posted

You are reading this latest preprint version

DDX20 interlinking with SNRPE endows MYC-driven triple-negative breast cancer therapeutic vulnerability in a splicing-independent manner

Status:

Version 1

Abstract

Figures

Background

Materials And Methods

Results

Discussion

Conclusions

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1