The nuclear receptor ERβ engages AGO2 in regulation of gene transcription, RNA splicing and RISC loading
- Roberta Tarallo†1,
- Giorgio Giurato†1, 2,
- Giuseppina Bruno1,
- Maria Ravo1, 2,
- Francesca Rizzo1,
- Annamaria Salvati1,
- Luca Ricciardi1,
- Giovanna Marchese2,
- Angela Cordella3,
- Teresa Rocco2,
- Valerio Gigantino1,
- Biancamaria Pierri1,
- Giovanni Cimmino4,
- Luciano Milanesi5,
- Concetta Ambrosino6, 7,
- Tuula A. Nyman8,
- Giovanni Nassa1Email author and
- Alessandro Weisz1Email authorView ORCID ID profile
© The Author(s). 2017
Received: 9 July 2017
Accepted: 20 September 2017
Published: 6 October 2017
The RNA-binding protein Argonaute 2 (AGO2) is a key effector of RNA-silencing pathways It exerts a pivotal role in microRNA maturation and activity and can modulate chromatin remodeling, transcriptional gene regulation and RNA splicing. Estrogen receptor beta (ERβ) is endowed with oncosuppressive activities, antagonizing hormone-induced carcinogenesis and inhibiting growth and oncogenic functions in luminal-like breast cancers (BCs), where its expression correlates with a better prognosis of the disease.
Applying interaction proteomics coupled to mass spectrometry to characterize nuclear factors cooperating with ERβ in gene regulation, we identify AGO2 as a novel partner of ERβ in human BC cells. ERβ–AGO2 association was confirmed in vitro and in vivo in both the nucleus and cytoplasm and is shown to be RNA-mediated. ChIP-Seq demonstrates AGO2 association with a large number of ERβ binding sites, and total and nascent RNA-Seq in ERβ + vs ERβ − cells, and before and after AGO2 knock-down in ERβ + cells, reveals a widespread involvement of this factor in ERβ-mediated regulation of gene transcription rate and RNA splicing. Moreover, isolation and sequencing by RIP-Seq of ERβ-associated long and small RNAs in the cytoplasm suggests involvement of the nuclear receptor in RISC loading, indicating that it may also be able to directly control mRNA translation efficiency and stability.
These results demonstrate that AGO2 can act as a pleiotropic functional partner of ERβ, indicating that both factors are endowed with multiple roles in the control of key cellular functions.
The argonaute protein AGO2 is a RNA-binding protein primarily known for its functions in the cytoplasm, where it is a major component of the RNA-induced silencing complex (RISC). Indeed, this factor controls miRNA maturation and is involved in target recognition by small non-coding RNAs, thereby leading to mRNA degradation or translation inhibition in post-transcriptional gene silencing [1–3]. The role of AGO2 in the composition of the miRNA machinery and the regulation of miRNA target stability and translation is well documented, among others, in breast cancer (BC) cells [4, 5]. On the other hand, AGO2 also acts in the nucleus, where it has been recently implicated in key events in several species, including mammals, such as transcriptional gene silencing (TGS) mediated by miRNAs [6–9], and it is involved in chromatin remodeling  and alternative RNA splicing  via RNA Pol II processivity slowdown and/or splicing factor recruitment . Recent results demonstrated that this protein can shuttle between the cytoplasm and nucleus, and that its subcellular distribution is context-dependent . Nucleocytoplasmic shuttling is a specific property also of estrogen receptor β (ERβ) [14, 15], a member of the nuclear receptor superfamily of transcriptional regulators  that shows oncosuppressive activities in BC and other cancers. In BC, where AGO2 has been shown to be associated with tumor progression , ERβ inhibits cancer cell proliferation and tumor growth and its expression has been found to correlate with a better prognosis of the disease . Furthermore, ERβ shows additive effects with anti-estrogens in promotion of apoptotic cell death and cell cycle inhibition [19, 20], and for this reason has been proposed as a marker of tumor responsiveness to endocrine therapy [21, 22]. Although this receptor can bind estrogenic compounds, thereby exerting a modulatory role on the functions of the oncogenic ERα, the other estrogen receptor subtype active in cancer cells, by dimerizing with it and thereby modifying its activity on target genes [16, 23], in the absence of ligand it exhibits significant effects in BC cells , including, among others, miRNA-mediated post-transcriptional regulation of the BC cell proteome . Physiologically, the presence of unliganded ERβ is a typical condition during specific phases of the menstrual cycle, before puberty, and in post-menopausal women, when this receptor might compensate for the absence of circulating hormones with regard to cell functions.
We show here that expression of unliganded ERβ in luminal-like BC MCF-7 cells induces profound effects on the cell transcriptome, represented by changes in both RNA expression and splicing. To elucidate the molecular bases of these actions, we applied interaction proteomics coupled to mass spectrometry (MS) to identify ERβ-interacting proteins in BC cell nuclei. AGO2 was among 277 new molecular partners of the receptor identified using this approach. Interestingly, a comparison between this protein dataset and datasets related to AGO2-interacting proteins present in public databases revealed a number of molecular partners in common between the two factors, indicating that they share a sizeable amount of functions in the nucleus, comprising also RNA processing and splicing. Based on these results, we investigated in depth the functional significance of ERβ–AGO2 interaction, identifying a dual role of the association between AGO2 and ERβ in BC cells in the nucleus and the cytoplasm, for quantitative and qualitative regulation of gene expression at both the transcriptional and post-transcriptional level.
In vivo binding of ERβ to the luminal-like BC cell genome and effects on gene expression
When considering, instead, ERβ-responsive genes comprising one or more receptor binding sites within the whole transcription unit, the number increases to 1752, including 476 (27%) whose RNA transcripts undergo alternative splicing in the presence of ERβ with patterns identical to those detectable on the whole transcriptome (Additional file 6: Figure S2). Since transcription factors, including nuclear receptors and, in particular, estrogen receptors [36–39], are known to regulate their target genes also through long-range chromatin looping interactions, the high number of ERβ binding sites mapped here suggests that a much higher number of these ERβ-dependent gene responses and splicing events are likely due to a direct effect of the receptor.
Mapping the nuclear interactome of unliganded ERβ identified AGO2 as a novel molecular partner of the receptor in BC cells
Upon agonist ligand binding, ERβ can dimerize with ERα in the nucleus , a condition that could affect AGO2 binding. However, AGO2 is found associated with ERβ also in the presence of 17β-estradiol (E2; Additional file 6: Figure S6a), while under the same condition it did not bind TAP-tagged ERα (Ct-ERα; Additional file 6: Figure S6b), demonstrating a estrogen receptor subtype-specific ability to associate with AGO2. This result indicates a selective role of the complex(es) comprising the two proteins in ERβ-specific functions in BC cells.
Considering the role exerted by AGO2 in the ERβ interactome, the nature of its interaction with the receptor, i.e., primary or mediated by additional factor(s), was further evaluated. A yeast two-hybrid assay performed using AGO2 fused to the LexA DNA binding domain as ‘bait’ and ERβ fused to the Gal4 activation domain as ‘prey’ failed to demonstrate direct association between the two proteins (data not shown), suggesting that other molecules could be involved in the interaction. For this reason, starting from the assumption that AGO2 is an RNA-binding protein, we investigated whether RNA could represent the bridging factor between the two proteins. To evaluate this possibility, cytoplasmic and nuclear protein extracts from Ct-ERβ cells were treated with RNAse A for different times before ERβ pull-down by IgG binding, followed by immunodetection of the two proteins in the immununoprecipitates, as described by Höck and colleagues . Results showed a strong reduction of ERβ and AGO2 association already 1 h after RNAse treatment, indicating that association between the two factors is indirect and likely to be mediated by one or more RNAs in both the nucleus and cytoplasm (Fig. 5b).
AGO2 binding to the BC cell genome in proximity of ERβ
AGO2 cooperates with unliganded ERβ to modulate transcription rate and RNA maturation
We then investigated the role of ERβ–AGO2 association in co-transcriptional pre-mRNA splicing. The coupling between transcription and splicing in eukaryotes is well known but the mechanisms that drive it are still unclear, although some evidence points to a kinetic and functional coupling between the two events that determine spliceosome assembly and pre-mRNA loading during transcription [62, 63]. Considering the involvement of ERβ in the control of the basic events of transcription and the identification, among ERβ-associated proteins, of several splicing factors associated also with nuclear AGO2 , we searched for evidence of RNA maturation and intron retention rate in the nascent-Seq datasets. By analyzing the data with the same procedure described for nascent transcript analysis, we investigated the global splicing events, in particular the intron retention coefficient, to verify the existence and nature of RNAs whose maturation may be modulated by ERβ–AGO2 functional cooperation during transcription. To quantify co-transcriptional splicing, we adopted the intron retention metric developed by Khodor et al. . In detail, to account for the variability due to imbalance among exons of different length, intron retention was calculated as the ratio between the read number/base pair of a given intron with respect to the same of all exons of the gene. In this way, we identified more than 11,500 splicing events (FDR ≤ 0.05, t-test) modulated by ERβ, with 5360 intron retention events being significantly affected in ERβ + cells compared to wild-type MCF-7 (Fig. 8c; Additional file 13: Table S10a), suggesting that ERβ may be directly involved also in this process in BC cells, as already demonstrated for AGO2. Subsequently, we measured 23,362 events (FDR ≤ 0.05, t-test) influenced only in ERβ + cells by AGO2 silencing (Fig. 8d; Additional file 13: Table S10b), a result obtained after filtering out the events observed also in ERβ − cells following AGO2 ‘knock-down’. By comparing these two datasets, and considering the 5360 introns influenced by ERβ, we highlighted several splicing events modulated by both ERβ and AGO2 (Fig. 8e). A stronger inhibition of the splicing efficiency upon AGO2 depletion was observed for ERβ-induced intron splicing, where the effect of the receptor was reverted in 78% of the cases showing an intron retention coefficient < −2 and in 52% of the cases where this coefficient was between −2 and −1.5. A similar, but less pronounced effect of AGO2 silencing was also observed for ERβ-mediated intron retention, with increased splicing efficiency in 55% of the cases when the intron retention coefficient was > 2 and of 37% when it was between 1.5 and 2. Taken together, these data indicate that ERβ and AGO2 cooperate in modulation of a sizeable amount of co-transcriptional splicing events in luminal-like BC cells, and that their functional association may be important to either promote or reduce the rate of co-transcriptional maturation of their target transcripts. This is further supported by the fact that only 8% of the 513 ERβ-dependent intron retention events occurring in the 99 genes showing ERβ–AGO2 complex binding were still detectable following AGO2 silencing, indicating that the large majority of them are AGO2-dependent.
Considering that ERβ–AGO2 co-occupancy occurs at 858 sites in chromatin, and the fact that the chromatin-bound nuclear receptor can exert transcriptional effects also through long-range chromatin looping, the results reported above strongly suggest a functional role of the cooperation between ERβ and AGO2 on direct regulation of gene transcription and co-transcriptional RNA splicing in BC cells. This is further supported by the fact that the ERβ interactome of BC cell nuclei includes several transcriptional co-regulators and components of the RNA splicing machinery (Fig. 3) and the evidence that AGO2 plays a central role in assembly and/or stability of the nuclear ERβ interactome (Fig. 4).
The AGO2–ERβ complex associates with long and small RNAs in BC cells
Given the extent of ERβ–AGO2 interaction in the extranuclear compartment observed in vivo by PLA (Additional file 6: Figure S5b) and in vitro by co-immunoprecipitation (Fig. 5) and the role of AGO2 as a major component of RISC [44, 45], we investigated the significance of this interaction in the cytoplasm.
When combined with the presence of RISC loading factors in the same ERβ complexes, the fact that most of these RNAs are present in low amounts in the cell suggests the involvement of ERβ in inducing recruitment of miRNAs and selected target mRNAs by the RISC loading machinery for destabilization of the latter.
The estrogen receptors ERα and ERβ are directly involved in carcinogenesis and tumor progression in multiple neoplasms of the female genital tract, and ERα was the first molecule amenable to drug targeting in BC, where its presence in cancer cells is still one of the main markers for identification of patients that will benefit from endocrine therapy. For a long time ERα was considered the only estrogen receptor in mammals, but a second one, termed ERβ, was subsequently discovered and found to play important roles in breast and other cancers . ERβ shows 55% identity with ERα in its ligand-binding domain and approximately 97% similarity in the DNA-binding domain (DBD). Reflecting the high degree of similarity in their DBDs, in chromatin both receptors target predominantly the same conserved estrogen response element (ERE; 5′-GGTCAnnnTGACC-3′) as either homodimers or α/β heterodimers . ERβ binds 17β-estradiol (E2) with relatively low affinity compared to ERα, but, contrary to ERα, shows potent effects also in the absence of ligands , like other members of the nuclear receptor superfamily of homeostatic regulators.
ERβ is expressed in normal mammary epithelial cells and in a fraction of BCs, showing decreased expression in cancer compared with benign tumors or normal tissues, suggesting that a reduction of this receptor in cancer cells could represent a critical stage in hormone-dependent tumor progression . Interestingly, Förster et al.  reported that ERβ null mice show impairment of pregnancy-induced terminal differentiation of the mammary gland, suggesting that this receptor subtype is required for normal development of this organ. When combined, these findings led to the hypothesis that ERβ might act as oncosuppressor in certain target tissues, including mammary epithelia, by interfering with the tumor promoting actions of estrogen via ERα and of other carcinogenic stimuli and by controlling genetic programs for cell differentiation and proliferation. This was further supported by the observation that mice lacking ERβ display multi-focal hyperplasia in prostate and bladder . For all the reasons stated above, understanding the molecular mechanisms of ERβ actions is a critical issue in cancer, in particular in BC biology. By interaction proteomics we identified molecular partners of both receptors in the nucleus of cells exposed to agonist and antagonist ligands [40–43, 69], and characterized the effects of unliganded ERβ in BC cells, demonstrating its significant effects on cell proliferation, miRNA expression, and the cell proteome .
In this study, we demonstrated that unliganded ERβ binds to the BC cell genome and induces reprogramming of the cell transcriptome, promoting also alternative splicing of a sizeable number of RNAs transcribed from its target genes. To understand the molecular determinants of these effects, we applied interaction proteomics and identified a large set of ligand-free ERβ-interacting proteins in the cell nucleus. The functions of several of the proteins reveal how this receptor can control key processes in BC cells, including gene transcription and RNA splicing and turnover. Among the molecular partners of ERβ, our attention was caught by AGO2, for the basic functions this protein exerts not only on miRNA biogenesis and actions but also on gene regulation. We thus investigated in detail the functional significance of the protein complex(es) containing AGO2 and ERβ, since we observed that several ERβ-interacting proteins were known to be also AGO2 interactors. These include, together with factors involved in RNA biogenesis, splicing, and maturation, also pleiotropic factors controlling key functions in the cell, such as CNOT1 (CCR4-NOT transcription complex subunit 1), a scaffolding component of the major effector complex of miRNA-mediated gene silencing CCR4-NOT, which associates with the ATP-dependent RNA helicase DDX6 (another interactor in common between ERβ and AGO2) to exert this function ; the metastasis-associated protein MTA2 (Metastasis associated 1 family member 2), a member of the tumor-associated family of transcriptional regulators and central component of nucleosome remodeling and histone deacetylation complexes , shown to be involved in both development and metastasis of a wide spectrum of cancers, including in particular hormone-independent BCs ; ADAR (Adenosine deaminase, RNA-specific), an RNA-editing enzyme specifically active in BC, where it has been shown to regulate cell proliferation and apoptosis ; COPA (Coatomer protein complex subunit alpha), a component of the coatomer complex of secretory vesicles involved in ER–Golgi transport whose mutation and inactivation have been recently shown to cause growth inhibition and apoptosis in cancer cells  and an autosomal dominant immune dysregulatory disease ; the human homolog of NOP56 (Nucleolar protein 56/NOP56 ribonucleoprotein), a core component of the C/D box snoRNP complex that controls ribosome biogenesis by regulating pre-rRNA processing and shows dynamic subcellular redistribution in response to growth conditions and nutrient availability .
Searching for the biological significance of ERβ–AGO2 association, transcriptional co-regulation of genes mediated by the joint action of the two proteins was demonstrated by the identification of several genomic regions occupied by both ERβ and AGO2 and by cooperation between the two proteins in modulation of transcription rate and co-transcriptional pre-mRNA splicing. Interestingly, AGO2 binding to the genome appears quite different in ERβ + compared to ERβ − cells (Fig. 6). This result, which was reproducible in independent experiments, suggests that the nuclear receptor can induce re-positioning of the argonaute protein within chromatin. Since AGO2 is not a DNA-binding protein and, therefore, its association with the genome is mediated by other factors, it is conceivable to assume that ERβ might influence the cellular levels of some of these factors, or their ability to bind DNA. Supporting the first possibility, we observed that the mRNAs encoding MGA, FOXP1, and GMEB2 are up-regulated in ERβ + cells, while those for ARID5A, GMEB1, NFATC2, NFAT5, TEAD1, and STAT4 are down-regulated. The binding matrixes for all these factors were significantly enriched within the AGO2 sites mapped here (Figs. 6 and 7). Furthermore, it is also possible that AGO2 tethering to the genome is mediated, in some instances, by RNAs whose expression is controlled by the receptor, which as shown here induces a profound effect on the cell transcriptome (Fig. 1; Additional file 1).
A set of 153 ERβ-responsive genes showing co-occupancy by ERβ and AGO2 on defined sites in their transcription units was identified, including 77 whose transcription rate was significantly affected (|FC| 1.5) in ERβ + cells upon AGO2 silencing. Functional analysis highlighted that these genes are involved in processes such as cell growth and proliferation, death and survival, or motility. Considering the overall involvement of these genes on cellular pathways, Gαq and phospholipase C signaling were significantly activated, while protein kinase A and B signaling may be inhibited. Both these pathways are tightly related to cancer progression and apoptosis. In particular, it has been demonstrated that G protein-coupled receptors are involved in BC progression and that Gq signaling promotes cancer cell apoptosis through phospholipase C [77–79]. On the other hand, protein kinase A signaling has been shown to promote mammary tumorigenesis  and to determine ERα repositioning at promoters and tamoxifen resistance ; its inhibition by ERβ–AGO2 cooperation may thus negatively affect cancer cell proliferation and survival.
In addition to the genes for which we could demonstrate both binding of the AGO2–ERβ nuclear complex and transcriptional regulation, we detected several others that are influenced in their transcription and/or maturation rate. Since it has been reported that AGO2 association with chromatin induces the formation of heterochromatin mediated by siRNAs in mammalian cells, probably determining slowdown of RNA polymerase II and alternative splicing events [12, 82], its association with ERβ may give rise to significant effects on gene activity via different mechanisms. Indeed, considering the capability of ERs to mediate transcription through long-range chromatin interactions , and the fact that AGO2 has been shown to co-localize with the insulator factor CTCF , known to mediate chromatin looping , association between AGO2 and ERβ may control gene activity also when occurring at a distance from the targeted transcription units. This could explain, at least in part, their massive effects shown here on gene transcription in BC cells. On the other hand, regulation of genome activity by the combined action of AGO2 and ERβ could occur via at least two, independent and not mutually exclusive, events. On one hand, binding to the genome of the complex(es) comprising the transcription factor and the argonaute protein together with other protein(s) and/or RNA(s) determines modulation of target gene expression. On the other, ERβ and AGO2 may bind nascent transcripts and modulate pre-mRNA splicing by recruitment and association with splicing factors. Indeed, we observed several such factors in common between the ERβ interactome identified here and the AGO2-associated proteins described by Ameyar-Zazoua et al. .
Association between AGO2 and ERβ also occurs, both in vivo and in vitro, in the cytoplasm, where isolation of ERβ-bound RNAs and miRNAs suggests that the receptor may assist the argonaute protein in the loading of specific miRNA–mRNA molecules in RISC, thus contributing also to post-transcriptional regulation of gene expression. Interestingly, we identified 868 RNAs and 18 miRNAs specifically associated with ERβ. Notably, computational analysis revealed that miRNA–mRNA molecules bound to ERβ are implicated in Wnt and cadherin signaling pathways. The first has been found dysregulated in BC [86, 87] and associated with metastasis in ‘triple negative’ tumors , while the second is tightly correlated to the Wnt, E-cadherin, and N-cadherin pathway, contributing significantly to epithelial–mesenchymal transition and metastasis . A negative effect of ERβ on translation and/or stability of the mRNAs encoding these factors could thus also be part of its activity as an oncosuppressor, contributing to the better prognosis of ERβ-expressing tumors. Combined with the relationships between AGO2 and tumorigenesis and cancer progression , the results reported here open new avenues for understanding the actions, and resulting effects, of ERβ and AGO2 in cancer cells.
Finally, interaction of AGO2 with ERβ appears to be indirect, since yeast two-hybrid assays failed to demonstrate direct association between the two proteins and, more important, in vitro RNAse A digestion of both nuclear and cytosolic extracts strongly reduced co-immunoprecipitation of the two proteins, suggesting that this interaction may require one or more RNAs. To our knowledge, ERβ binding to RNA has not been described previously; however, this is well known for other nuclear receptors, such as the androgen receptor  and ERα , where a novel RNA binding domain in the N-terminus of the protein has been identified . Attempts to identify the RNA(s) involved in ERβ–AGO2 complex formation and/or stability have so far been unsuccessful, but a preliminary computational prediction, performed on RNAs specifically binding to ERβ and AGO2 in the cytosolic compartment, suggested that long noncoding RNAs could be likely candidates as bridging molecules (data not shown). Understanding this aspect will need further investigations that go beyond the scope of the present study.
The results of this study demonstrate that AGO2 and ERβ can physically and functionally associate, both in the nucleus and the cytoplasm, in complex(es) comprising also several other proteins and RNAs. The final biological outcome of such association appears to depend upon the sum of different variables, including transcriptional, splicing, and post-transcriptional events and, possibly, the specific cellular context. These findings provide new leads toward understanding the oncosuppressive role of ERβ via regulation of gene transcription, RNA maturation, and post-transcriptional control of RNA activity, and the consequences of the loss of this protein in transformed cells. Demonstration of the general importance of these results, obtained here in a cellular model of ERβ + BC, for the control of cellular functions and its derangement during carcinogenesis and tumor progression will require, however, further validation in less artificial conditions, in particular in vivo animal models, patient-derived xenografts and tumor biopsies.
Stable clones expressing ERβ tagged with TAP-tag at either the C-terminus (Ct-ERβ) or N-terminus (Nt-ERβ) and TAP-tagged ERα were obtained from human breast cancer MCF-7 Tet-Off cells (ER-alpha positive; ATCC, catalog number HTB-22) as previously described [40, 69]. For generation of ERβ tagged inducible clones, the human full-length cDNA clone pCMV6-ESR2 (RC218519) encoding human ESR2 was purchased from Origene. ESR2 sequence, including the Myc and Flag tags, was subcloned into the BamHI and EcoRI sites of pTRE-Tight expression vector (Clontech). All cell lines were propagated in Dulbecco’s modified Eagle medium (DMEM; Sigma-Aldrich) supplemented with 10% FBS (HyClone) and antibiotics: 100 U/ml penicillin, 100 mg/ml streptomycin, 250 ng/ml Amfotericin-B. Steroid deprivation (starvation) was performed by culturing in DMEM without phenol red and 5% dextran coated charcoal stripped serum (DCC-FBS) for 5 days. Cell lines were authenticated by short tandem repeat (STR) profiling and routinely tested for Mycoplasma contamination with MycoAlert mycoplasma detection kit (Lonza).
RNA extraction sequencing and data analysis
Total RNA was extracted from ERβ + and ERβ − (Ct-ERα and/or wild-type) MCF-7 cells using the standard RNA extraction method with TRIzol (Life Technologies). Before use, the RNA concentration in each sample was assayed with a ND-2000c spectrophotometer (Thermoscientific) and its quality and integrity assessed with the Agilent 2100 Bioanalyzer with Agilent RNA 6000 nano kit (Agilent Technologies). For RNA sequencing experiments, indexed libraries were prepared using 1 μg of total RNA as starting material, with a TruSeq Stranded Total RNA Sample Prep Kit (Illumina Inc.). Libraries were sequenced (paired-end, 2 × 100 cycles) at a concentration of 8 pM/lane on the HiSeq 2500 platform (Illumina Inc.). The raw sequence files generated (.fastq files) underwent quality control analysis using FASTQC (http://www.bioinformatics.babraham.ac.uk/projects/fastqc/) and quality-checked reads were then aligned to the human genome (assembly hg19) using TopHat version 2.0.10  with the standard parameters. The expression value of each mRNA was normalized to FPKM (fragments per kilobase of exon model per million of sequenced reads) as computed by Cufflink . Differentially expressed mRNAs were identified using DESeq2 . Firstly, gene annotation was obtained for all known genes in the human genome, as provided by Ensemble (GRCh37; https://support.illumina.com/sequencing/sequencing_software/igenome.html). Using the reads mapped to the genome, we computed the number of reads mapping to each transcript with HTSeq-count . A given mRNA was considered expressed when detected by at least ≥ 10 reads. The raw read counts were then used as input to DESeq for calculation of normalized signal for each transcript in the samples. Differential expression was reported as fold change |1.5| along with associated adjusted p values (FDR ≤ 0.05), computed according to Benjamini–Hochberg. Alternative splicing data analysis was performed as described previously . Raw RNA sequencing data are deposited in the EBI ArrayExpress database (http://www.ebi.ac.uk/arrayexpress) with accession number E-MTAB-4363.
Chromatin immunoprecipitation, sequencing, and data analysis
C-TAP-ERβ and MCF-7 control cells were hormone-deprived for 5 days. For each assay, a total of about 15 × 106 cells were fixed, lysed to isolate nuclei, sonicated, and diluted as described by Schmidt et al. , with minor modifications. An aliquot of nuclear extract was taken as input. For ERβ pull-down, chromatin samples were incubated, as described earlier , at 4 °C for 3 h with 40 μl of IgG Sepharose 6 fast Flow (GE Healthcare Bio-Science AB) properly equilibrated in Poly-Prep chromatography columns (0.84 cm, Bio-Rad), according to the manufacturer’s instructions. For AGO2 immunoprecipitation, chromatin samples were incubated at 4 °C overnight with 40 μl of pre-blocked magnetic beads (Dynabeads, Thermofisher) conjugated with 1 μg of mouse monoclonal anti-AGO2/eIF2C2 (ab57113, Abcam). As negative control for these experiments, chromatin samples were also incubated overnight with 1 μg of mouse monoclonal anti-Flag M2 affinity purified (F1804, Sigma-Aldrich). Bead washing, elution, reverse crosslinking and DNA extraction were then performed as described . The size distribution of each ChIP DNA sample was assessed by running a 1 μl aliquot on an Agilent High Sensitivity DNA chip using an Agilent Technologies 2100 Bioanalyzer (Agilent Technologies). The concentration of each DNA sample was determined by using a Quant-IT DNA Assay Kit-High Sensitivity and a Qubit Fluorometer (Life Technologies). Purified ChIP DNA (10 ng) was used as the starting material for sequencing library preparation from three independent ChIP experiments. Indexed triplicate libraries were prepared with a TruSeq ChIP Sample Prep Kit (Illumina Inc.) and were sequenced (single read, 1 × 50 cycles) on a NextSeq 500 (Illumina Inc.).
Read alignment and quality control of ChIP-seq data
The raw sequence files generated (.fastq) underwent quality control analysis using FASTQC (http://www.bioinformatics.babraham.ac.uk/projects/fastqc/). Reads were aligned to the reference human genome assembly (hg19) using bowtie , allowing up to one mismatch and considering uniquely mappable reads. Duplicated reads were removed using Picard tools v 2.9.0 (MarkDuplicates; https://broadinstitute.github.io/picard).
For each biological replicate and corresponding control samples, peak calling was performed using MACS2  with p value set to 0.05. The peaks obtained for each biological replicate were combined using MuSERA  with the following parameters: replicate type, biological; Ts, 1E-08; Tw, 1E-04; γ, 1E-08; Benjamini–Hochberg false discovery rate (α), 0.005, using the lowest p value when multiple regions from a sample intersected with the region of another sample and considering peaks common to at least two replicates (C:2). The annotation of peaks to the nearest gene was performed combining the information obtained using the annotatePeaks.pl function of HOMER  and the Annotation and Statistics of Genomatix Software suite. Comparison, integration, and quantification were performed using seqMINER . Over-represented sequence motifs for known transcription factors, according to motif descriptors in the JASPAR database, were determined using PScan-ChIP . Only over-represented motifs with p value ≤ 1E-10 were considered.
De novo motif discovery
The predicted sequences of ERβ and Ago2 binding sites were extracted and used for de novo motif discovery using the RSAT peak motifs method with default parameters  and Meme-ChIP . For ERβ binding sites, the ERE binding motif was searched with MatInspector application , using a core similarity threshold of 0.75 and a matrix similarity threshold of Optimal −0.02.
Binding site statistics
The overlap between ERβ and Ago2 binding sites was calculated using bedtools intersect . The significance of the overlaps was assessed using the poverlap tool (https://github.com/brentp/poverlap) by performing 100,000 simulations and allowing shuffling of both datasets. The significance of overlaps of ERβ with different genomic regions (3′ UTR, 5′ UTR, intergenic, exonic, intronic, promoter, and TSS) was assessed using Genomic Association Test (GAT)  with 10,000 simulations. In each case considered, the distance to the TSS was computed using ChIPseek . Raw ChIP-Seq data have been deposited in the EBI ArrayExpress database (http://www.ebi.ac.uk/arrayexpress) with accession number E-MTAB-4359.
Tandem affinity purification
Nuclear extraction and tandem affinity purification from C-TAP-ERβ and control (wild-type MCF-7) cells were performed as previously described [40, 41, 69]. Partially purified samples, coming from the first purification step, were then subjected to mass spectrometry analysis for protein identification.
Nano LC-MS/MS and data analysis
Three biological replicates of partially purified samples from Ct-ERβ and control MCF-7 cells were separated on SDS-PAGE and visualized with silver-staining. After separation, SDS-PAGE lanes were sliced into six pieces, and the proteins were in-gel digested with trypsin into peptides and analyzed by LC-MS/MS as previously described . MS data were acquired using Analyst QS 2.0 software. The information-dependent acquisition method consisted of a 0.5 s TOF-MS survey scan of m/z 400–1400. From every survey scan two most abundant ions with charge states +2 to +4 were selected for product ion scans. Once an ion was selected for MS/MS fragmentation, it was put on an exclusion list for 60 s. LC-MS/MS data from the biological replicates were combined and searched against SwissProt 2010 (517,802 sequences, 182,492,287 residues; human, 20,283 sequences) for control (TAP-only) and SwissProt 2010 (523,151 sequences, 184,678,199 residues; human, 20,259 sequences) for Ct-ERβ samples. The search criteria for Mascot searches were: trypsin digestion with one missed cleavage allowed, carbamidomethyl (C) as fixed modification and oxidation (M), phospho (ST), phospho (Y) as variable modifications. For the LC-MS/MS spectra the maximum precursor ion mass tolerance was 50 ppm and MS/MS fragment ion mass tolerance 0.2 Da, and a peptide charge state of +1, +2, or +3 was used. All of the reported protein identifications were statistically significant (p < 0.05). To eliminate the redundancy of proteins that appear in the database under different names and accession numbers, the single protein member with the highest protein score (top rank) was selected from multiprotein families for the identification results. Protein reported as Ct-ERβ molecular interacting partners were selected by filtering them against the proteins identified in negative control after quality assessment of the identification peptides. The Mascot search results, including peptide sequences identifying Ct-ERβ interacting proteins, are reported in Additional file 7: Table S4 (Mascot search results sheets).
For experiments performed in the presence or absence of AGO2 silencing, three biological replicates of partially purified samples from Ct-ERβ for each of the two conditions were analyzed. The proteins were precipitated with 10% TCA in acetone solution and dissolved in 40 μL 0.2% ProteaseMAX™ Surfactant, Trypsin Enhancer (Promega) in 50 mM NH4HCO3 followed by protein reduction, alkylation, and in-solution digestion with trypsin (Promega), performed overnight at 37 °C. Peptides were desalted and concentrated before mass spectrometry by the STAGE-TIP method, using a C18 resin disk (3 M Empore). The peptides were eluted twice with 0.1% TFA/50% ACN, dried, and solubilized in 7 μL 0.1% TFA for mass spectrometry analysis. Each peptide mixture was analyzed on an Easy nLC1000 nano-LC system connected to a quadrupole Orbitrap mass spectrometer (QExactive Plus, ThermoElectron) equipped with a nanoelectrospray ion source (EasySpray/Thermo). For the liquid chromatography separation of the peptides, an EasySpray column capillary of 50 cm bed length (C18, 2 μm beads, 100 Å, 75 μm inner diameter, Thermo) was employed. The flow rate was 300 nL/min, and the peptides were eluted with a 2–30% gradient of solvent B in 120 min. Solvent A was aqueous 0.1% formic acid and solvent B 100% acetonitrile/0.1% formic acid. The data-dependent acquisition automatically switched between MS and MS/MS mode. Survey full scan MS spectra were acquired from a mass-to-charge ration (m/z) of 400 to 1200 with the resolution R = 70,000 at m/z 200 after accumulation to a target of 3,000,000 ions in the quadruple. For MS/MS, the ten most abundant multiple-charged ions were selected for fragmentation on the high-energy collision dissociation (HCD) cell at a target value of 100,000 charges or maximum acquisition time of 100 ms. The MS/MS scans were collected at a resolution of 17,500. Target ions already selected for MS/MS were dynamically excluded for 30 s. The resulting MS raw files were submitted to MaxQuant software version 188.8.131.52 for protein identification using the Andromeda search engine. Carbamidomethyl (C) was set as a fixed modification and protein N-acetylation and methionine oxidation were set as variable modifications. First search peptide tolerance of 20 ppm and main search error 4.5 ppm were used. Trypsin without the proline restriction enzyme option was used, with two allowed miscleavages. The minimal unique + razor peptide number was set to 1, and the allowed FDR was 0.01 (1%) for peptide and protein identification. Label-free quantification (LFQ) was employed with default settings. The SwissProt human database (August 2016 release, with 154,660 entries) was used for the database searches. Known contaminants as provided by MaxQuant and identified in the samples were excluded from further analysis. LFQ intensities were used for differential expression analysis. Protein LFQ values were further normalized by ESR2 LFQ value in each replicate for each dataset. Then, to identify statistically modulated proteins a two-sample t-test statistical analysis with a permutation based FDR cut-off of 0.01 was performed. All the protein identification and quantification data are reported in Additional file 8: Table S5a–d.
The mass spectrometry proteomics data have been deposited in the ProteomeXchange Consortium via the PRIDE  partner repository with the dataset identifier PXD006280.
Nuclear protein extraction and co-immunoprecipitation
Nuclear protein extracts were prepared, as described , from inducible MCF-7 Tet-On cells expressing Myc-Flag-ERβ treated or not with doxycycline (2 μg/ml) for 24 h. To immunoprecipitate AGO2, 1 mg of nuclear proteins was incubated overnight at 4 °C with 2 μg of mouse monoclonal anti-AGO2/eIF2C2 (ab57113, Abcam) and then at 4 °C for 1 h with 35 μl of equilibrated slurry Protein A/G Plus-Agarose (sc-2003, Santa Cruz Biotechnology). To immunoprecipitate myc-flag-tagged ERβ, the same amount of nuclear proteins was incubated for 2 h at 4 °C with 35 μl of equilibrated slurry EZview Red Anti-c-Myc Affinity Gel (E6654, Sigma Aldrich). After binding, the beads were sequentially washed with IPP150 buffer (7.14 mM HEPES pH 7.5, 8.92% glycerol, 150 mM NaCl, 0.54 mM MgCl2, 0.07 mM EDTA pH 8, 1× protease inhibitors) and wash buffer (50 mM Tris-HCl pH 7.6, 150 mM NaCl, 1× protease inhibitors). To elute ERβ-immunoprecipitated samples from the beads, an elution at 4 °C for 30 min was performed using c-Myc Peptide (M 2435, Sigma Aldrich).
MCF-7 Tet-On cells stably expressing tet-inducible Myc-Flag-ERβ were seeded on microscope glass slides and starved for 3 days before treatment with doxycyline for 24 h. Cells were then fixed with 4% paraformaldehyde for 15 min, washed with PBS-Tween three times and permeabilized with 0.1% Triton X-100 in PBS. After washing with PBS and blocking with 0.5% BSA, slides were incubated with mouse anti-Myc (clone 4A6, Merck Millipore, 1:200) and rabbit anti-AGO2 (07-590, Millipore, 1:66), incubated and washed for three times each with 0.5% BSA and then PBS, before incubation with Alexa Fluor 488 goat anti-mouse IgG (Thermofisher) and Cy3 anti-rabbit (Jackson Immuno Research) secondary antibodies. BSA and PBS washes were repeated and cells were covered with mounting medium containing 4′,6-diamidino-2-phenylindole (DAPI 1:20,000) and imaged with a confocal microscope (Leica DM6000 B). Images were processed with ImageJ software (https://imagej.net).
Proximity ligation assay
MCF-7 cells were plated on microscope glass slides and after 5 days of starvation, transiently transfected with plasmids expressing either Myc-tagged AGO2, Flag-tagged ERβ, or Flag-tagged ERα. Non-transfected cells were used as control. Cells were washed three times in ice-cold PBS and fixed by incubating them in 4% paraformaldehyde for 20 min under gentle agitation in the dark. After three washes with PBS, cells were permeabilized with 0.2% Triton X100 for 5 min under gentle agitation and then washed again. A proximity ligation assay was performed following the manufacturer’s instructions. In detail, fixed and permeabilized cells were blocked in a pre-heated humidity chamber for 30 min at 37 °C with one drop of blocking solution per 1 cm2. Then, primary antibodies were added (rabbit anti-Flag Tag, F7425 ad mouse anti-Myc Tag: clone 4A6, Merck Millipore) and incubated for 1 h at 37 °C in a pre-heated humidity chamber. Slides were washed twice for 5 min in wash buffer A in a staining jar with gentle orbital shaking and then incubated with PLA probes (Mouse ± for the detection of exogenous AGO2, Rabbit ± for the detection of ERβ or ERα and Mouse − and Rabbit + for the detection of AGO2/ER interactions) in a pre-heated humidity chamber for 1 h at 37 °C. After two other washes with wash buffer A, a ligation reaction was performed by adding the ligase to the slides (1:40 dilution of the stock) and incubating them in a humidity chamber for 30 min at 37 °C. Slides were washed twice with wash buffer A for 2 min under gentle agitation and then the amplification-polymerase solution was added to the cells and left to act in a pre-heated humidity chamber for 100 min at 37 °C. Two last final wash steps were performed, submerging slides twice in wash buffer B for 10 min and then in 0.01× wash buffer B for 1 min. The slides were than dried in the dark, prepared for imaging by adding Duolink II Mounting Medium with DAPI, and visualized using a confocal microscope (Leica DM6000 B).
SDS-PAGE and western blot analyses were performed using standard protocols. The following primary antibodies were used: rabbit anti-TAP (CAB1001, Thermo Scientific-Pierce), anti-Myc Tag clone 4A6 (05-724, Millipore), rabbit anti ERα (sc-543, Santa Cruz Biotechnology), mouse anti-AGO2/eIF2C2 (ab57113, Abcam), rabbit polyclonal to FXR1 (ab50841, Abcam), rabbit plyclonal to integrin beta 4 binding protein (EIF6; ab77298, Abcam), anti-PRPF8 antiboby (ab79237, Abcam), mouse monoclonal anti-AGO1 clone 4G7-E12 (MABE143, Millipore), mouse anti-β-actin (A1978, Sigma Aldrich), mouse monoclonal to Dicer (ab14601, Abcam), anti-TRBP2 (H-57; sc-292550, Santa Cruz Biotechnology).
For nascent-Seq experiments, C-TAP-ERβ and MCF-7 control cells were starved for 5 days and then AGO2 knock-down was performed using a combination of three pLKO.1 plasmid vectors expressing shRNAs (Sigma Aldrich: TRCN0000007864; TRCN0000007867; TRCN0000011203) targeting the AGO2 transcript (GenBankTM accession number NM_012154) in different regions. AGO2 silencing was conducted by co-transfecting C-TAP-ERβ and control cells with shRNA vectors, using Lipofectamine 2000 (Life Technologies), for 48 h. The transfection medium was replaced with fresh culturing medium 6 h after treatment. Non-transfected and transfected cells with pLKO.1-puro Non-Target shRNA Control Plasmid DNA (Sigma-Aldrich) were used as control. For TAP/MS after AGO2 silencing, hormone-deprived Ct-ERβ cells were transfected with SMARTvector human lentiviral shRNA pooled libraries (Dharmacon) for 72 h. Western blotting was performed to verify the level of ‘knock-down’ of the target protein.
Nascent RNA isolation, sequencing, and data analysis
Nascent RNA was extracted from each sample as described by Khodor et al. . In brief, following TRIzol (Life Techonolgies) addition, samples were incubated at 65 °C to dissolve DNA-Histone-Pol II-RNA pellets and RNA was extracted following the manufacturer’s protocol. For sequencing, indexed libraries were prepared using 1 μg of Nascent RNA as starting material, with TruSeq Stranded Total RNA Sample Prep Kit (Illumina Inc.). Libraries were sequenced (paired-end, 2 × 100 cycles) at a concentration of 8 pM/lane on the HiSeq 2500 platform (Illumina Inc.) .
Alignment to the human genome
Raw sequence files (.fastq files) underwent quality control analysis using FASTQC (http://www.bioinformatics.babraham.ac.uk/projects/fastqc/) and the quality checked reads were then aligned to the human genome (assembly hg19) using TopHat version 2.0.10 , according to the criteria used by Menet et al. .
Quantification of gene signal
Quantification of nascent RNA was done as in Menet et al. . Differentially expressed nascent RNAs were identified using DESeq2 . The differential expression was reported as fold change |1.5| along with associated adjusted p values (FDR ≤ 0.05) computed according to Benjamini–Hochberg.
Intron retention determination
Before proceeding with intronic quantification, we extracted intronic intervals as described by St Laurent et al. , while intron retention was computed as described by Khodor et al. . The statistical significance of intron retention events observed between the several conditions was assessed using t-test (FDR < 0.05). Raw data are deposited in the EBI ArrayExpress database (http://www.ebi.ac.uk/arrayexpress) with accession number EMTAB-4368.
RNA immunoprecipitation, sequencing, and data analysis
Cells were lysed with polysome lysis buffer, as described by Keene et al. . An aliquot of whole-cell extract (10% of total) was taken as input. For ERβ immunoprecipitation, samples were incubated at 4 °C for 3 h with 50 μl of IgG Sepharose 6 fast Flow (GE Healthcare Bio-Science AB) pre-treated with NT2 buffer supplemented with 5% BSA. After binding, the isolation of RNA co-precipitated with ERβ was carried out by adding TRIzol (Life Technologies) directly to the washed beads, following the manufacturer’s instructions. For RNA-Seq analyses, indexed sequencing libraries were prepared starting from 1 μg of RNA input and 300 ng of RNA immunoprecipitated, pooling three independent experiments (biological replicates) and using TruSeq Stranded Total RNA. For miRNA-Seq experiments, libraries were generated from 120 ng of the same pooled RNA using TruSeq Small RNA Sample Prep Kits (Illumina Inc.). Libraries were sequenced (single read 1 × 50 cycles and 2 × 100 cycles for miRNA- and RNA-Seq experiments, respectively) on a HiSeq 2500 (Illumina Inc.). Data analysis was performed as follows.
Raw sequence files (.fastq files) underwent quality control analysis using FASTQC (http://www.bioinformatics.babraham.ac.uk/projects/fastqc/) and the quality checked reads were then aligned to the human genome (assembly hg19) using TopHat version 2.0.10 . HTSeq-count  was used to compute gene-level read counts.
The read counts obtained were used as input to DESeq2  to perform enrichment analysis. RNAs showing enrichment factor (EF) > 1 and adjusted p value ≤ 0.05 were considered for further analysis. To define enriched RNAs in Ct-ERβ IP versus input RNAs, we applied a more stringent analysis: firstly, we selected the RNAs showing an EF more than 75th percentile of its distribution, and subsequently we compared these RNAs with those identified comparing wild-type IPβ versus input RNAs. Hence, we selected those RNAs specific for the Ct-ERβ IPβ group, and those that, when compared to wild type, showed a ratio (Ct-ERβ EF/wild-type EF) ≥ 4, or showing a negative EF in wild-type IPβ vs input. Small RNA-Seq data were analyzed using iSmaRT  with standard parameters, using miRBase v20 as reference track. miRNAs showing EF > 1 and p value ≤ 0.05 were considered for further analysis. To select ERβ-specific enriched miRNAs, those with EF > 1.5 were considered and the EF in ERβ + cells was compared with the same in wild-type cells. miRNAs showing a ratio between the two conditions (i.e., Ct-ERβ EF/wild-type EF) ≥ 2 or showing a negative EF in wild-type IPβ vs input were selected. The different classes of small RNAs obtained in IPβ in Ct-ERβ cells were assessed using sRNABench . Classification of enriched RNAs was performed using the "Gene biotype" term in ENSEMBL using a.gtf file downloaded from Genecode (http://www.gencodegenes.org/#).
Functional and pathway analyses
Functional and interaction network analysis of ERβ-associated proteins was performed with the FunRich tool  according to the user manual. The lists of transcripts were analyzed using Ingenuity Pathway Analysis Software (IPA, Ingenuity® Systems, www.ingenuity.com). It refers to a proprietary knowledge base (Ingenuity Pathways Knowledge Base) in which cellular molecules, biological interactions, and functional properties are annotated. IPA Functional Analysis on “molecular and cellular functions” category and Canonical Pathway investigation were carried out, calculating the likelihood that the association between our RNA dataset and a specific function or pathway is due to random choice, and it is expressed as a p value calculated using the right-tailed Fisher exact test. The activation z-score is used to infer likely activation states of enriched pathways, based on comparison with a model that assigns random regulation directions. Finally, the “microRNA Target Filter” IPA module was used to provide insights into the biological effects of microRNAs, using miRNA–mRNA interactions from TarBase and miRecords, as well as predicted miRNA–mRNA interactions from TargetScan examining miRNA–mRNA pairings in the pathways of interest. Finally, a network representing miRNA–RNA target interaction was created using Cytoscape .
GN was supported by a ‘Mario e Valeria Rindi’ fellowship of the Italian Foundation for Cancer Research.
Work supported by: Italian Association for Cancer Research (grants IG-17426), Italian Ministry for Education, University and Research (grant FIRB RBFR12W5V5_003 to RT), Italian Ministry of Health (Young Researcher grants GR-2011-02347781 to GN and GR-2011-02350476 to MR), University of Salerno (Fondi FARB 2015-2016) and CNR (Flagship Project InterOmics). We also acknowledge ELIXIR-IIB (http://elixir-italy.org/), the Italian Node of the European ELIXIR infrastructure (https://elixir-europe.org/), for the computational power support provided.
Availability of data and materials
The sequencing datasets generated and analyzed during the current study are available in the EBI ArrayExpress database repository (http://www.ebi.ac.uk/arrayexpress) with accession numbers E-MTAB-4363, E-MTAB-4359, and EMTAB-4368. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE partner repository with the dataset identifier PXD006280.
All authors participated in conception and design of the study. RT, GB, GN, MR, FR, GM, AC, TR, GC, VG, and BP performed in vitro experimental work and RNA sequencing. AS and LR performed in vivo experimental work. TN and GN performed the proteomics analyses. GG performed the statistical and bioinformatics analyses. RT, GG, GN, CA, LM, and AW coordinated and finalized figure preparation, manuscript drafting, and revision. All authors read and approved the final manuscript.
Ethics approval and consent to participate
Consent for publication
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Meister G, Landthaler M, Patkaniowska A, Dorsett Y, Teng G, Tuschl T. Human Argonaute2 mediates RNA cleavage targeted by miRNAs and siRNAs. Mol Cell. 2004;15:185–97.PubMedView ArticleGoogle Scholar
- Rand TA, Ginalski K, Grishin NV, Wang X. Biochemical identification of Argonaute 2 as the sole protein required for RNA-induced silencing complex activity. Proc Natl Acad Sci U S A. 2004;101:14385–9.PubMedPubMed CentralView ArticleGoogle Scholar
- Liu J, Carmell MA, Rivas FV, Marsden CG, Thomson JM, Song JJ, et al. Argonaute2 is the catalytic engine of mammalian RNAi. Science. 2004;305:1437–41.PubMedView ArticleGoogle Scholar
- Kwon SY, Lee JH, Kim B, Park JW, Kwon TK, Kang SH, Kim S. Complexity in regulation of microRNA machinery components in invasive breast carcinoma. Pathol Oncol Res. 2014;20:697–705.PubMedView ArticleGoogle Scholar
- Fan M, Krutilina R, Sun J, Sethuraman A, Yang CH, Wu ZH, et al. Comprehensive analysis of microRNA (miRNA) targets in breast cancer cells. J Biol Chem. 2013;288:27480–93.PubMedPubMed CentralView ArticleGoogle Scholar
- Morris KV. RNA-mediated transcriptional gene silencing in human cells. Curr Top Microbiol Immunol. 2008;320:211–24.PubMedGoogle Scholar
- Janowski BA, Huffman KE, Schwartz JC, Ram R, Nordsell R, Shames DS, et al. Involvement of AGO1 and AGO2 in mammalian transcriptional silencing. Nat Struct Mol Biol. 2006;13:787–92.PubMedView ArticleGoogle Scholar
- Kim DH, Saetrom P, Snove Jr O, Rossi JJ. MicroRNA-directed transcriptional gene silencing in mammalian cells. Proc Natl Acad Sci U S A. 2008;105:16230–5.PubMedPubMed CentralView ArticleGoogle Scholar
- Benhamed M, Herbig U, Ye T, Dejean A, Bischof O. Senescence is an endogenous trigger for microRNA-directed transcriptional gene silencing in human cells. Nat Cell Biol. 2012;14:266–75.PubMedPubMed CentralView ArticleGoogle Scholar
- Carissimi C, Laudadio I, Cipolletta E, Gioiosa S, Mihailovich M, Bonaldi T, et al. ARGONAUTE2 cooperates with SWI/SNF complex to determine nucleosome occupancy at human Transcription Start Sites. Nucleic Acids Res. 2015;43:1498–512.PubMedPubMed CentralView ArticleGoogle Scholar
- Taliaferro JM, Aspden JL, Bradley T, Marwha D, Blanchette M, Rio DC. Two new and distinct roles for Drosophila Argonaute-2 in the nucleus: alternative pre-mRNA splicing and transcriptional repression. Genes Dev. 2013;27:378–89.PubMedPubMed CentralView ArticleGoogle Scholar
- Ameyar-Zazoua M, Rachez C, Souidi M, Robin P, Fritsch L, Young R, et al. Argonaute proteins couple chromatin silencing to alternative splicing. Nat Struct Mol Biol. 2012;19:998–1004.PubMedView ArticleGoogle Scholar
- Sharma NR, Wang X, Majerciak V, Ajiro M, Kruhlak M, Meyers C, Zheng ZM. Cell-type and tissue context-dependent nuclear distribution of human Ago2. J Biol Chem. 2016;291:2302–9.Google Scholar
- Damdimopoulos AE, Spyrou G, Gustafsson JA. Ligands differentially modify the nuclear mobility of estrogen receptors alpha and beta. Endocrinology. 2008;149:339–45.PubMedView ArticleGoogle Scholar
- Maruvada P, Baumann CT, Hager GL, Yen PM. Dynamic shuttling and intranuclear mobility of nuclear hormone receptors. J Biol Chem. 2003;278:12425–32.PubMedView ArticleGoogle Scholar
- Thomas C, Gustafsson JA. The different roles of ER subtypes in cancer biology and therapy. Nat Rev Cancer. 2011;11:597–608.PubMedView ArticleGoogle Scholar
- Adams BD, Claffey KP, White BA. Argonaute-2 expression is regulated by epidermal growth factor receptor and mitogen-activated protein kinase signaling and correlates with a transformed phenotype in breast cancer cells. Endocrinology. 2009;150:14–23.PubMedView ArticleGoogle Scholar
- Leung YK, Lee MT, Lam HM, Tarapore P, Ho SM. Estrogen receptor-beta and breast cancer: translating biology into clinical practice. Steroids. 2012;77:727–37.PubMedPubMed CentralView ArticleGoogle Scholar
- Hodges-Gallagher L, Valentine CD, El Bader S, Kushner PJ. Estrogen receptor beta increases the efficacy of antiestrogens by effects on apoptosis and cell cycling in breast cancer cells. Breast Cancer Res Treat. 2008;109:241–50.PubMedView ArticleGoogle Scholar
- Murphy LC, Peng B, Lewis A, Davie JR, Leygue E, Kemp A, et al. Inducible upregulation of oestrogen receptor-beta1 affects oestrogen and tamoxifen responsiveness in MCF7 human breast cancer cells. J Mol Endocrinol. 2005;34:553–66.PubMedView ArticleGoogle Scholar
- Huang B, Warner M, Gustafsson JA. Estrogen receptors in breast carcinogenesis and endocrine therapy. Mol Cell Endocrinol. 2015;418(Pt 3):240–4.PubMedView ArticleGoogle Scholar
- Madeira M, Mattar A, Logullo AF, Soares FA, Gebrim LH. Estrogen receptor alpha/beta ratio and estrogen receptor beta as predictors of endocrine therapy responsiveness-a randomized neoadjuvant trial comparison between anastrozole and tamoxifen for the treatment of postmenopausal breast cancer. BMC Cancer. 2013;13:425.PubMedPubMed CentralView ArticleGoogle Scholar
- Grober OM, Mutarelli M, Giurato G, Ravo M, Cicatiello L, De Filippo MR, et al. Global analysis of estrogen receptor beta binding to breast cancer cell genome reveals an extensive interplay with estrogen receptor alpha for target gene regulation. BMC Genomics. 2011;12:36.PubMedPubMed CentralView ArticleGoogle Scholar
- Vivar OI, Zhao X, Saunier EF, Griffin C, Mayba OS, Tagliaferri M, et al. Estrogen receptor beta binds to and regulates three distinct classes of target genes. J Biol Chem. 2010;285:22059–66.PubMedPubMed CentralView ArticleGoogle Scholar
- Nassa G, Tarallo R, Giurato G, De Filippo MR, Ravo M, Rizzo F, et al. Post-transcriptional regulation of human breast cancer cell proteome by unliganded estrogen receptor beta via microRNAs. Mol Cell Proteomics. 2014;13:1076–90.PubMedPubMed CentralView ArticleGoogle Scholar
- Zhao C, Lam EW, Sunters A, Enmark E, De Bella MT, Coombes RC, et al. Expression of estrogen receptor beta isoforms in normal breast epithelial cells and breast cancer: regulation by methylation. Oncogene. 2003;22:7600–6.PubMedView ArticleGoogle Scholar
- Chang EC, Frasor J, Komm B, Katzenellenbogen BS. Impact of estrogen receptor beta on gene networks regulated by estrogen receptor alpha in breast cancer cells. Endocrinology. 2006;147:4831–42.PubMedView ArticleGoogle Scholar
- Lin CY, Strom A, Li Kong S, Kietz S, Thomsen JS, Tee JB, et al. Inhibitory effects of estrogen receptor beta on specific hormone-responsive gene expression and association with disease outcome in primary breast cancer. Breast Cancer Res. 2007;9:R25.PubMedPubMed CentralView ArticleGoogle Scholar
- Williams C, Edvardsson K, Lewandowski SA, Strom A, Gustafsson JA. A genome-wide study of the repressive effects of estrogen receptor beta on estrogen receptor alpha signaling in breast cancer cells. Oncogene. 2008;27:1019–32.PubMedView ArticleGoogle Scholar
- Dago DN, Scafoglio C, Rinaldi A, Memoli D, Giurato G, Nassa G, et al. Estrogen receptor beta impacts hormone-induced alternative mRNA splicing in breast cancer cells. BMC Genomics. 2015;16:367.PubMedPubMed CentralView ArticleGoogle Scholar
- Shen S, Park JW, Huang J, Dittmar KA, Lu ZX, Zhou Q, et al. MATS: a Bayesian framework for flexible detection of differential alternative splicing from RNA-Seq data. Nucleic Acids Res. 2016;291:2302–9.Google Scholar
- Paris O, Ferraro L, Grober OM, Ravo M, De Filippo MR, Giurato G, et al. Direct regulation of microRNA biogenesis and expression by estrogen receptor beta in hormone-responsive breast cancer. Oncogene. 2012;31:4196–206.PubMedView ArticleGoogle Scholar
- Zhang Y, Liu T, Meyer CA, Eeckhoute J, Johnson DS, Bernstein BE, et al. Model-based analysis of ChIP-Seq (MACS). Genome Biol. 2008;9:R137.PubMedPubMed CentralView ArticleGoogle Scholar
- Jalili V, Matteucci M, Morelli MJ, Masseroli M. MuSERA: Multiple Sample Enriched Region Assessment. Brief Bioinform. 2017;18:367–81.Google Scholar
- Bado I, Nikolos F, Rajapaksa G, Gustafsson JA, Thomas C. ERbeta decreases the invasiveness of triple-negative breast cancer cells by regulating mutant p53 oncogenic function. Oncotarget. 2016;7:13599–611.PubMedPubMed CentralView ArticleGoogle Scholar
- Carter D, Chakalova L, Osborne CS, Dai YF, Fraser P. Long-range chromatin regulatory interactions in vivo. Nat Genet. 2002;32:623–6.PubMedView ArticleGoogle Scholar
- Fullwood MJ, Liu MH, Pan YF, Liu J, Xu H, Mohamed YB, et al. An oestrogen-receptor-alpha-bound human chromatin interactome. Nature. 2009;462:58–64.PubMedPubMed CentralView ArticleGoogle Scholar
- Korkmaz G, Lopes R, Ugalde AP, Nevedomskaya E, Han R, Myacheva K, et al. Functional genetic screens for enhancer elements in the human genome using CRISPR-Cas9. Nat Biotechnol. 2016;34:192–8.PubMedView ArticleGoogle Scholar
- Bourgo RJ, Singhal H, Greene GL. Capture of associated targets on chromatin links long-distance chromatin looping to transcriptional coordination. Nat Commun. 2016;7:12893.PubMedPubMed CentralView ArticleGoogle Scholar
- Nassa G, Tarallo R, Ambrosino C, Bamundo A, Ferraro L, Paris O, et al. A large set of estrogen receptor beta-interacting proteins identified by tandem affinity purification in hormone-responsive human breast cancer cell nuclei. Proteomics. 2011;11:159–65.PubMedView ArticleGoogle Scholar
- Tarallo R, Bamundo A, Nassa G, Nola E, Paris O, Ambrosino C, et al. Identification of proteins associated with ligand-activated estrogen receptor alpha in human breast cancer cell nuclei by tandem affinity purification and nano LC-MS/MS. Proteomics. 2011;11:172–9.PubMedView ArticleGoogle Scholar
- Cirillo F, Nassa G, Tarallo R, Stellato C, De Filippo MR, Ambrosino C, et al. Molecular mechanisms of selective estrogen receptor modulator activity in human breast cancer cells: identification of novel nuclear cofactors of antiestrogen-ERalpha complexes by interaction proteomics. J Proteome Res. 2013;12:421–31.PubMedView ArticleGoogle Scholar
- Stellato C, Nassa G, Tarallo R, Giurato G, Ravo M, Rizzo F, et al. Identification of cytoplasmic proteins interacting with unliganded estrogen receptor alpha and beta in human breast cancer cells. Proteomics. 2015;15:1801–7.PubMedView ArticleGoogle Scholar
- Hutvagner G, Simard MJ. Argonaute proteins: key players in RNA silencing. Nat Rev Mol Cell Biol. 2008;9:22–32.PubMedView ArticleGoogle Scholar
- Hock J, Meister G. The Argonaute protein family. Genome Biol. 2008;9:210.PubMedPubMed CentralView ArticleGoogle Scholar
- Kalathur RK, Pinto JP, Hernandez-Prieto MA, Machado RS, Almeida D, Chaurasia G, Futschik ME. UniHI 7: an enhanced database for retrieval and interactive analysis of human molecular interaction networks. Nucleic Acids Res. 2014;42:D408–14.PubMedView ArticleGoogle Scholar
- Cowley MJ, Pinese M, Kassahn KS, Waddell N, Pearson JV, Grimmond SM, et al. PINA v2.0: mining interactome modules. Nucleic Acids Res. 2012;40:D862–5.PubMedView ArticleGoogle Scholar
- Hock J, Weinmann L, Ender C, Rudel S, Kremmer E, Raabe M, et al. Proteomic and functional analysis of Argonaute-containing mRNA-protein complexes in human cells. EMBO Rep. 2007;8:1052–60.PubMedPubMed CentralView ArticleGoogle Scholar
- Cho S, Park JS, Kang YK. AGO2 and SETDB1 cooperate in promoter-targeted transcriptional silencing of the androgen receptor gene. Nucleic Acids Res. 2014;42:13545–56.PubMedPubMed CentralView ArticleGoogle Scholar
- Clocchiatti A, Di Giorgio E, Viviani G, Streuli C, Sgorbissa A, Picco R, et al. The MEF2-HDAC axis controls proliferation of mammary epithelial cells and acini formation in vitro. J Cell Sci. 2015;128:3961–76.PubMedView ArticleGoogle Scholar
- Zhou Y, Huang T, Cheng AS, Yu J, Kang W, To KF. The TEAD family and its oncogenic role in promoting tumorigenesis. Int J Mol Sci. 2016;17(1):138.Google Scholar
- Zang H, Li N, Pan Y, Hao J. Identification of upstream transcription factors (TFs) for expression signature genes in breast cancer. Gynecol Endocrinol. 2017;33:193–98.PubMedView ArticleGoogle Scholar
- Quang CT, Leboucher S, Passaro D, Fuhrmann L, Nourieh M, Vincent-Salomon A, Ghysdael J. The calcineurin/NFAT pathway is activated in diagnostic breast cancer cases and is essential to survival and metastasis of mammary cancer cells. Cell Death Dis. 2015;6:e1658.PubMedView ArticleGoogle Scholar
- Jauliac S, Lopez-Rodriguez C, Shaw LM, Brown LF, Rao A, Toker A. The role of NFAT transcription factors in integrin-mediated carcinoma invasion. Nat Cell Biol. 2002;4:540–4.PubMedView ArticleGoogle Scholar
- Watson CJ. Stat transcription factors in mammary gland development and tumorigenesis. J Mammary Gland Biol Neoplasia. 2001;6:115–27.PubMedView ArticleGoogle Scholar
- Svingen T, Tonissen KF. Altered HOX gene expression in human skin and breast cancer cells. Cancer Biol Ther. 2003;2:518–23.PubMedView ArticleGoogle Scholar
- Dravis C, Spike BT, Harrell JC, Johns C, Trejo CL, Southard-Smith EM, et al. Sox10 regulates stem/progenitor and mesenchymal cell states in mammary epithelial cells. Cell Rep. 2015;12:2035–48.PubMedPubMed CentralView ArticleGoogle Scholar
- Heger A, Webber C, Goodson M, Ponting CP, Lunter G. GAT: a simulation framework for testing the association of genomic intervals. Bioinformatics. 2013;29:2046–8.PubMedPubMed CentralView ArticleGoogle Scholar
- Wang X, Guda C. Integrative exploration of genomic profiles for triple negative breast cancer identifies potential drug targets. Medicine (Baltimore). 2016;95:e4321.View ArticleGoogle Scholar
- Elsarraj HS, Hong Y, Valdez KE, Michaels W, Hook M, Smith WP, et al. Expression profiling of in vivo ductal carcinoma in situ progression models identified B cell lymphoma-9 as a molecular driver of breast cancer invasion. Breast Cancer Res. 2015;17:128.PubMedPubMed CentralView ArticleGoogle Scholar
- Zatula N, Wiese M, Bunzendahl J, Birchmeier W, Perske C, Bleckmann A, Brembeck FH. The BCL9-2 proto-oncogene governs estrogen receptor alpha expression in breast tumorigenesis. Oncotarget. 2014;5:6770–87.PubMedPubMed CentralView ArticleGoogle Scholar
- Ip JY, Schmidt D, Pan Q, Ramani AK, Fraser AG, Odom DT, Blencowe BJ. Global impact of RNA polymerase II elongation inhibition on alternative splicing regulation. Genome Res. 2011;21:390–401.PubMedPubMed CentralView ArticleGoogle Scholar
- Das R, Dufu K, Romney B, Feldt M, Elenko M, Reed R. Functional coupling of RNAP II transcription to spliceosome assembly. Genes Dev. 2006;20:1100–9.PubMedPubMed CentralView ArticleGoogle Scholar
- Khodor YL, Rodriguez J, Abruzzi KC, Tang CH, Marr 2nd MT, Rosbash M. Nascent-seq indicates widespread cotranscriptional pre-mRNA splicing in Drosophila. Genes Dev. 2011;25:2502–12.PubMedPubMed CentralView ArticleGoogle Scholar
- Siraj AK, Beg S, Jehan Z, Prabhakaran S, Ahmed M, RH A, et al. ALK alteration is a frequent event in aggressive breast cancers. Breast Cancer Res. 2015;17:127.PubMedPubMed CentralView ArticleGoogle Scholar
- Bardin A, Boulle N, Lazennec G, Vignon F, Pujol P. Loss of ERbeta expression as a common step in estrogen-dependent tumor progression. Endocr Relat Cancer. 2004;11:537–51.PubMedPubMed CentralView ArticleGoogle Scholar
- Forster C, Makela S, Warri A, Kietz S, Becker D, Hultenby K, et al. Involvement of estrogen receptor beta in terminal differentiation of mammary gland epithelium. Proc Natl Acad Sci U S A. 2002;99:15578–83.PubMedPubMed CentralView ArticleGoogle Scholar
- Krege JH, Hodgin JB, Couse JF, Enmark E, Warner M, Mahler JF, et al. Generation and reproductive phenotypes of mice lacking estrogen receptor beta. Proc Natl Acad Sci U S A. 1998;95:15677–82.PubMedPubMed CentralView ArticleGoogle Scholar
- Ambrosino C, Tarallo R, Bamundo A, Cuomo D, Franci G, Nassa G, et al. Identification of a hormone-regulated dynamic nuclear actin network associated with estrogen receptor alpha in human breast cancer cell nuclei. Mol Cell Proteomics. 2010;9:1352–67.PubMedPubMed CentralView ArticleGoogle Scholar
- Chen Y, Boland A, Kuzuoglu-Ozturk D, Bawankar P, Loh B, Chang CT, et al. A DDX6-CNOT1 complex and W-binding pockets in CNOT9 reveal direct links between miRNA target recognition and silencing. Mol Cell. 2014;54:737–50.PubMedView ArticleGoogle Scholar
- Covington KR, Fuqua SA. Role of MTA2 in human cancer. Cancer Metastasis Rev. 2014;33:921–8.PubMedPubMed CentralView ArticleGoogle Scholar
- Ning Z, Gan J, Chen C, Zhang D, Zhang H. Molecular functions and significance of the MTA family in hormone-independent cancer. Cancer Metastasis Rev. 2014;33:901–19.PubMedView ArticleGoogle Scholar
- Fumagalli D, Gacquer D, Rothe F, Lefort A, Libert F, Brown D, et al. Principles governing A-to-I RNA editing in the breast cancer transcriptome. Cell Rep. 2015;13:277–89.PubMedPubMed CentralView ArticleGoogle Scholar
- Sudo H, Tsuji AB, Sugyo A, Kohda M, Sogawa C, Yoshida C, et al. Knockdown of COPA, identified by loss-of-function screen, induces apoptosis and suppresses tumor growth in mesothelioma mouse model. Genomics. 2010;95:210–6.PubMedView ArticleGoogle Scholar
- Watkin LB, Jessen B, Wiszniewski W, Vece TJ, Jan M, Sha Y, et al. COPA mutations impair ER-Golgi transport and cause hereditary autoimmune-mediated lung disease and arthritis. Nat Genet. 2015;47:654–60.PubMedPubMed CentralView ArticleGoogle Scholar
- Kakihara Y, Makhnevych T, Zhao L, Tang W, Houry WA. Nutritional status modulates box C/D snoRNP biogenesis by regulated subcellular relocalization of the R2TP complex. Genome Biol. 2014;15:404.PubMedPubMed CentralView ArticleGoogle Scholar
- Zuo H, Wong YH. Association of activated Galphaq to the tumor suppressor Fhit is enhanced by phospholipase Cbeta. BMC Cancer. 2015;15:775.PubMedPubMed CentralView ArticleGoogle Scholar
- Dorsam RT, Gutkind JS. G-protein-coupled receptors and cancer. Nat Rev Cancer. 2007;7:79–94.PubMedView ArticleGoogle Scholar
- Golebiewska U, Guo Y, Khalikaprasad N, Zurawsky C, Yerramilli VS, Scarlata S. gamma-Synuclein interacts with phospholipase Cbeta2 to modulate G protein activation. PLoS One. 2012;7:e41067.PubMedPubMed CentralView ArticleGoogle Scholar
- Beristain AG, Molyneux SD, Joshi PA, Pomroy NC, Di Grappa MA, Chang MC, et al. PKA signaling drives mammary tumorigenesis through Src. Oncogene. 2015;34:1160–73.PubMedView ArticleGoogle Scholar
- de Leeuw R, Flach K, Bentin Toaldo C, Alexi X, Canisius S, Neefjes J, et al. PKA phosphorylation redirects ERalpha to promoters of a unique gene set to induce tamoxifen resistance. Oncogene. 2013;32:3543–51.PubMedView ArticleGoogle Scholar
- Allo M, Buggiano V, Fededa JP, Petrillo E, Schor I, de la Mata M, et al. Control of alternative splicing through siRNA-mediated transcriptional gene silencing. Nat Struct Mol Biol. 2009;16:717–24.PubMedView ArticleGoogle Scholar
- Liu MH, Cheung E. Estrogen receptor-mediated long-range chromatin interactions and transcription in breast cancer. Mol Cell Endocrinol. 2014;382:624–32.PubMedView ArticleGoogle Scholar
- Moshkovich N, Nisha P, Boyle PJ, Thompson BA, Dale RK, Lei EP. RNAi-independent role for Argonaute2 in CTCF/CP190 chromatin insulator function. Genes Dev. 2011;25:1686–701.PubMedPubMed CentralView ArticleGoogle Scholar
- Holwerda SJ, de Laat W. CTCF: the protein, the binding partners, the binding sites and their chromatin loops. Philos Trans R Soc Lond B Biol Sci. 2013;368:20120369.PubMedPubMed CentralView ArticleGoogle Scholar
- Jang GB, Kim JY, Cho SD, Park KS, Jung JY, Lee HY, et al. Blockade of Wnt/beta-catenin signaling suppresses breast cancer metastasis by inhibiting CSC-like phenotype. Sci Rep. 2015;5:12465.PubMedPubMed CentralView ArticleGoogle Scholar
- Zardawi SJ, O'Toole SA, Sutherland RL, Musgrove EA. Dysregulation of Hedgehog, Wnt and Notch signalling pathways in breast cancer. Histol Histopathol. 2009;24:385–98.PubMedGoogle Scholar
- Dey N, Barwick BG, Moreno CS, Ordanic-Kodani M, Chen Z, Oprea-Ilies G, et al. Wnt signaling in triple negative breast cancer is associated with metastasis. BMC Cancer. 2013;13:537.PubMedPubMed CentralView ArticleGoogle Scholar
- Ashaie MA, Chowdhury EH. Cadherins: the superfamily critically involved in breast cancer. Curr Pharm Des. 2016;22:616–38.PubMedView ArticleGoogle Scholar
- Ye Z, Jin H, Qian Q. Argonaute 2: a novel rising star in cancer research. J Cancer. 2015;6:877–82.PubMedPubMed CentralView ArticleGoogle Scholar
- Yang L, Lin C, Jin C, Yang JC, Tanasa B, Li W, et al. lncRNA-dependent mechanisms of androgen-receptor-regulated gene activation programs. Nature. 2013;500:598–602.PubMedPubMed CentralView ArticleGoogle Scholar
- Lanz RB, McKenna NJ, Onate SA, Albrecht U, Wong J, Tsai SY, et al. A steroid receptor coactivator, SRA, functions as an RNA and is present in an SRC-1 complex. Cell. 1999;97:17–27.PubMedView ArticleGoogle Scholar
- Ghosh SK, Patton JR, Spanjaard RA. A small RNA derived from RNA coactivator SRA blocks steroid receptor signaling via inhibition of Pus1p-mediated pseudouridylation of SRA: evidence of a novel RNA binding domain in the N-terminus of steroid receptors. Biochemistry. 2012;51:8163–72.PubMedView ArticleGoogle Scholar
- Trapnell C, Pachter L, Salzberg SL. TopHat: discovering splice junctions with RNA-Seq. Bioinformatics. 2009;25:1105–11.PubMedPubMed CentralView ArticleGoogle Scholar
- Trapnell C, Roberts A, Goff L, Pertea G, Kim D, Kelley DR, et al. Differential gene and transcript expression analysis of RNA-seq experiments with TopHat and Cufflinks. Nat Protoc. 2012;7:562–78.PubMedPubMed CentralView ArticleGoogle Scholar
- Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15:550.PubMedPubMed CentralView ArticleGoogle Scholar
- Anders S, Huber W. Differential expression analysis for sequence count data. Genome Biol. 2010;11:R106.PubMedPubMed CentralView ArticleGoogle Scholar
- Schmidt D, Wilson MD, Spyrou C, Brown GD, Hadfield J, Odom DT. ChIP-seq: using high-throughput sequencing to discover protein-DNA interactions. Methods. 2009;48:240–8.PubMedPubMed CentralView ArticleGoogle Scholar
- Langmead B, Trapnell C, Pop M, Salzberg SL. Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome Biol. 2009;10:R25.PubMedPubMed CentralView ArticleGoogle Scholar
- Heinz S, Benner C, Spann N, Bertolino E, Lin YC, Laslo P, et al. Simple combinations of lineage-determining transcription factors prime cis-regulatory elements required for macrophage and B cell identities. Mol Cell. 2010;38:576–89.PubMedPubMed CentralView ArticleGoogle Scholar
- Ye T, Krebs AR, Choukrallah MA, Keime C, Plewniak F, Davidson I, Tora L. seqMINER: an integrated ChIP-seq data interpretation platform. Nucleic Acids Res. 2011;39:e35.PubMedView ArticleGoogle Scholar
- Zambelli F, Pesole G, Pavesi G. PscanChIP: Finding over-represented transcription factor-binding site motifs and their correlations in sequences from ChIP-Seq experiments. Nucleic Acids Res. 2013;41:W535–43.PubMedPubMed CentralView ArticleGoogle Scholar
- Medina-Rivera A, Defrance M, Sand O, Herrmann C, Castro-Mondragon JA, Delerce J, et al. RSAT 2015: regulatory sequence analysis tools. Nucleic Acids Res. 2015;43:W50–6.PubMedPubMed CentralView ArticleGoogle Scholar
- Machanick P, Bailey TL. MEME-ChIP: motif analysis of large DNA datasets. Bioinformatics. 2011;27:1696–7.PubMedPubMed CentralView ArticleGoogle Scholar
- Cartharius K, Frech K, Grote K, Klocke B, Haltmeier M, Klingenhoff A, et al. MatInspector and beyond: promoter analysis based on transcription factor binding sites. Bioinformatics. 2005;21:2933–42.PubMedView ArticleGoogle Scholar
- Quinlan AR. BEDTools: the Swiss-Army tool for genome feature analysis. Curr Protoc Bioinformatics. 2014;47:11. 12 1-34.PubMedPubMed CentralGoogle Scholar
- Chen TW, Li HP, Lee CC, Gan RC, Huang PJ, Wu TH, et al. ChIPseek, a web-based analysis tool for ChIP data. BMC Genomics. 2014;15:539.PubMedPubMed CentralView ArticleGoogle Scholar
- Vizcaino JA, Csordas A, Del-Toro N, Dianes JA, Griss J, Lavidas I, et al. 2016 update of the PRIDE database and its related tools. Nucleic Acids Res. 2016;44:11033.PubMedPubMed CentralView ArticleGoogle Scholar
- Menet JS, Rodriguez J, Abruzzi KC, Rosbash M. Nascent-Seq reveals novel features of mouse circadian transcriptional regulation. Elife. 2012;1:e00011.PubMedPubMed CentralView ArticleGoogle Scholar
- St Laurent G, Shtokalo D, Tackett MR, Yang Z, Eremina T, Wahlestedt C, et al. Intronic RNAs constitute the major fraction of the non-coding RNA in mammalian cells. BMC Genomics. 2012;13:504.PubMedPubMed CentralView ArticleGoogle Scholar
- Keene JD, Komisarow JM, Friedersdorf MB. RIP-Chip: the isolation and identification of mRNAs, microRNAs and protein components of ribonucleoprotein complexes from cell extracts. Nat Protoc. 2006;1:302–7.PubMedView ArticleGoogle Scholar
- Anders S, Pyl PT, Huber W. HTSeq--a Python framework to work with high-throughput sequencing data. Bioinformatics. 2015;31:166–9.PubMedView ArticleGoogle Scholar
- Panero R, Rinaldi A, Memoli D, Nassa G, Ravo M, Rizzo F, et al. iSmaRT: a toolkit for a comprehensive analysis of small RNA-Seq data. Bioinformatics. 2017;33:938–40.PubMedGoogle Scholar
- Rueda A, Barturen G, Lebron R, Gomez-Martin C, Alganza A, Oliver JL, Hackenberg M. sRNAtoolbox: an integrated collection of small RNA research tools. Nucleic Acids Res. 2015;43:W467–73.PubMedPubMed CentralView ArticleGoogle Scholar
- Pathan M, Keerthikumar S, Ang CS, Gangoda L, Quek CY, Williamson NA, et al. FunRich: An open access standalone functional enrichment and interaction network analysis tool. Proteomics. 2015;15:2597–601.PubMedView ArticleGoogle Scholar
- Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, et al. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res. 2003;13:2498–504.PubMedPubMed CentralView ArticleGoogle Scholar