MaTAR25 lncRNA regulates the Tensin1 gene to impact breast cancer progression

Abstract

Misregulation of long non-coding RNA (lncRNA) genes has been linked to a wide variety of cancer types. Here we report on Mammary Tumor Associated RNA 25 (MaTAR25), a nuclear enriched and chromatin associated lncRNA that plays a role in mammary tumor cell proliferation, migration, and invasion, both in vitro and in vivo. MaTAR25 functions by interacting with purine rich element binding protein B (PURB), and associating with a major downstream target gene Tensin1 (Tns1) to regulate its expression in trans. The Tns1 protein product is a critical component of focal adhesions linking signaling between the extracellular matrix and the actin cytoskeleton. Knockout of MaTAR25 results in down-regulation of Tns1 leading to a reorganization of the actin cytoskeleton, and a reduction of focal adhesions and microvilli. We identify LINC01271 as the human ortholog of MaTAR25, and importantly, increased expression of LINC01271 is associated with poor patient prognosis and metastasis. Our findings demonstrate that LINC01271 represents a potential therapeutic target to alter breast cancer progression.

Fig. 1. Characterization of MammaryTumor AssociatedRNA25 (MaTAR25).

a Representation of the MaTAR25 gene locus. MaTAR25 is an intergenic lncRNA gene located on mouse chromosome 2, and the MaTAR25 RNA transcript contains 2 exons and a poly (A) tail. b 5′ and 3′ rapid amplification of cDNA ends (RACE) was performed to identify the full-length MaTAR25 transcript. Three independent experiments were performed, and a representative gel image is shown. c The full-length MaTAR25 transcript was confirmed by northern blot analysis to be ~2000 nt. In total, 20 μg or 30 μg of total RNA samples extracted from MMTV-PyMT primary cells were electrophoresed on a 1% agarose gel and probed. Two independent experiments were performed, and a representative gel image is shown. d In vitro transcription and translation reactions were performed to confirm that MaTAR25 does not produce a peptide. The reaction products were loaded on a 4–20% gradient SDS-PAGE gel, and the signals were detected by HRP-conjugated streptavidin. Luciferase control DNA and Xenopus laevis Histone H2B (HISTH2B) expressing plasmids were used as positive controls and empty vector as a negative control. Three independent experiments were performed, and a representative gel image is shown. e Representative smRNA-FISH images showing localization of MaTAR25 RNA transcripts (red) within nuclei of MMTV-PyMT and MMTV-Neu-NDL primary cells. Scale bars are 10 μm.

Fig. 2. MaTAR25 knockout affects 4T1 cell viability, migration, and invasion in vitro; all of which can be rescued by ectopic expression of MaTAR25 in knockout cells.

a CRISPR/Cas9 was used to generate MaTAR25 KO clones in 4T1 cells. Pairs of sgRNAs were introduced targeting upstream and downstream of the transcription start site of MaTAR25, resulting in 390–620 bp genomic deletions over this region, and a Renilla Luciferase sgRNA was used as a negative control. Knockout clones were selected by genomic PCR and Sanger sequencing for homozygous genomic deletion. Control clones were selected by qRT-PCR for expressing similar levels of MaTAR25 as parental 4T1 cells. qRT-PCR (n = 2), and representative images of single smRNA-FISH (red: MaTAR25) are shown to confirm MaTAR25 KO. Scale bars are 5 μm. b 4T1 cells were seeded at the same cell density in 12-well tissue culture plates at day 0, and cell counting was performed at different time points. The mean cell numbers of three independent replicates of 4T1 control groups and MaTAR25 KO groups are shown ± SD (n = 3). *P < 0.05 (paired Student’s t test; two-tailed). c Live-cell tracking was performed over time to examine cell migration. Images were collected every 5 min for a total of 8 h and analyzed by CellTracker image-processing software. The mean relative migration distance (μm) of 4T1 control groups and MaTAR25 KO groups is shown ± SD (n = 3 independent replicates). *P < 0.05 (paired Student’s t test; two-tailed). d Twenty-four-well Boyden chamber invasion assay (24 h). The mean relative cell invasion of three independent replicates of 4T1 control groups and MaTAR25 KO groups is shown ± SD (n = 3). *P < 0.05 (paired Student’s t test; two-tailed). e Selected clones with ectopic expression of MaTAR25 or GFP were used as positive and negative controls to assess rescue in a cell viability assay. Data are presented as mean values ± SD (n = 3). *P < 0.05 (paired Student’s t test; two-tailed). f Cell invasion assay. The mean cell numbers and mean relative cell invasion of three independent replicates of 4T1 control1, MaTAR25 KO1, MaTAR25 KO1 with GFP expression, and MaTAR25 KO1 with MaTAR25 ectopic expression is shown ± SD (n = 3). *P < 0.05 (paired Student’s t test; two-tailed).

Fig. 3. MaTAR25 knockout impairs tumor growth and metastasis in vivo.

a 4T1 control or MaTAR25 KO cells were injected orthotopically into the mammary fat pad of female BALB/c mice. Primary tumors were measured every week over a period of 4 weeks, and the mean tumor volume of eight mice per group is shown ± SE. *P < 0.05 (paired Student’s t test; two-tailed). b Ten mice were sacrificed, and tumors were collected at day 28 to compare the tumor growth rate between the control group and MaTAR25 KO groups. Tumors derived from MaTAR25 KO cells showed a 56% reduction in tumor growth. The mean tumor wet weight is shown ± SE. *P < 0.05 (paired Student’s t test; two-tailed). c Female BALB/c mice were injected into the tail vein with 4T1 control1 or MaTAR25 KO cells. Mice were monitored every week and sacrificed at day 21. Mouse lungs were collected and imaged (left panel), and lung metastatic nodules were counted to compare the metastatic ability between the control group and MaTAR25 KO group (right panel). Mice injected with MaTAR25 KO cells exhibited a 62% reduction in lung metastatic nodules. Data are presented as mean values shown ± SD. *P < 0.05 (paired Student’s t test; two-tailed). d Schematic showing the approach for ASO-mediated knockdown of MaTAR25 in MMTV-Neu-NDL mice. Two independent MaTAR25 ASOs or a control scASO were used for subcutaneous injection. Primary tumors were measured twice per week, and the mean tumor volume of seven mice per group is shown ± SE. *P < 0.05 (paired Student’s t test; two-tailed). e Hematoxylin and eosin (H&E)-stained tumor images showing the different histological phenotypes between tumor samples from the scASO-injected group and MaTAR25 ASO-injected group. More than five tumor samples in each group were examined, and representative images are shown. Scale bars are 3 mm.

Fig. 4. MaTAR25 is a positive upstream regulator of Tns1.

a Cell fractionation was performed to isolate cytoplasmic, nucleoplasmic, and chromatin-associated RNA. qRT-PCR was used to determine the subcellular localization ratio of MaTAR25 transcripts. β-actin and Malat1 were used as marker RNAs for quality control. Data are presented as mean values ± SD (n = 3 independent experiments). b Schematic diagram showing the targeting of biotin-labeled oligonucleotides binding MaTAR25 transcripts for Chromatin Isolation by RNA Purification (ChIRP)-seq. Odd and even oligo pools (seven oligos in each pool) were used, and qRT-PCR was performed to check the RNA purification efficiency. Data are presented as mean values ± SD (n = 3 independent experiments). c Venn diagram showing differentially expressed genes in MaTAR25 KO cells identified from RNA-Seq overlapped with MaTAR25 ChIRP-seq data. The top candidate genes are listed. Statistics were determined using DESeq2, and a FDR-adjusted P value of <0.1 was set as a threshold for statistical significance. d Validation of Tns1 as a MaTAR25-targeted gene by qRT-PCR and immunoblotting in 4T1 control and MaTAR25 KO cells (n = 2 independent experiments). e The RNA expression level of Tns1 is rescued upon ectopic expression of MaTAR25 in MaTAR25 KO cells as determined by qRT-PCR (n = 2 independent experiments). f CRISPR/Cas9 targeting was used in 4T1 cells to generate Tns1 knockout clones. The upper panel shows expression levels of Tns1 in 4T1 control, Tns1 KO clone1, and Tns1 KO clone2 by immunoblotting. The lower panel shows the cell counting viability assay result of 4T1 control1, Tns1 KO1, and Tns1 KO2. Results are mean ± SD (n = 3 independent experiments). *P < 0.05 (paired Student’s t test; two-tailed). g Ectopic expression of Tns1 in MaTAR25 KO cells rescues the cell viability defect. The top panel shows expression levels of Tns1 in 4T1 control1, MaTAR25 KO1, MaTAR25 KO1 with Tns1 ectopic expression clone3 by qRT-PCR. Results are mean ± SD (n = 3 independent experiments). *P < 0.05 (paired Student’s t test; two-tailed). The bottom panel shows the cell counting viability assay results of 4T1 control1, MaTAR25 KO1, MaTAR25 KO1 with MaTAR25 ectopic expression, and MaTAR25 KO1 with Tns1 ectopic expression clone1-3. Results are mean ± SD (n = 2 independent experiments).

Fig. 5. MaTAR25 interacts with PURB to carry out its function.

a Scatterplot depicts the fold enrichment of protein candidates from isobaric tags for the relative and absolute quantitation (iTRAQ) analysis comparing two independent oligo pair sets targeting MaTAR25 RNA transcripts vs Ppib RNA transcripts. b Upper: immunoblot analysis of PURA and PURB following pulldown of MaTAR25 or Ppib from 4T1 cells. Lower: immunoblot analysis of PURB following the pulldown of MaTAR25 or Ppib from 4T1 cells or 4T1 MaTAR25 KO cells. More than three different experiments were performed, and representative images are shown. c MaTAR25, Ppib, and Gapdh transcripts were assessed by qRT-PCR in endogenous PURB, or IgG (negative control) immunoprecipitates from 4T1 cells. Fold enrichment of PURB associated RNA signal over IgG signal is calculated and data are presented as mean values ± SD (n = 3 independent experiments). *P < 0.05 (paired Student’s t test; two-tailed). Immunoblot analysis of PURB was performed as a control. d qRT-PCR analysis and immunoblotting of Tns1 expression in 4T1 cells following ectopic overexpression of PURB. The relative expression levels are shown as mean values ± SD (n = 2 independent experiments). e ChIP-qPCR analysis of PURB occupancy over the identified MaTAR25 targeting region and non-targeting region of the Tns1 DNA locus by ChIRP-seq analysis. ChIP-qPCR was performed in 4T1 control1 cells, 4T1 MaTAR25 KO1 cells, and upon ectopic expression of MaTAR25 in MaTAR25 KO1 cells. Primers for a MaTAR25 non-targeting region and the Gapdh TSS were used as negative controls. Bar graphs represent the mean ± SD (n = 2 independent experiments).

Fig. 6. Human LINC01271 is the human ortholog of MaTAR25.

a All potential human orthologs of MaTAR25 (hMaTAR25) were identified based on conservation of genomic location (synteny). RNA-seq data from The Cancer Genome Atlas (TCGA) was analyzed to evaluate the expression status of all potential hMaTAR25 candidates by comparison of 1128 TCGA breast tumor datasets to 113 normal breast tissue controls. Fold change and statistical significance were calculated using the Wilcoxon rank test; P values were adjusted using the Benjamini & Hochberg method by DESeq2. b Attempted rescue of 4T1 MaTAR25 KO cells upon ectopic expression of two transcript isoforms of LINC01270 (LINC01270.1, and LINC01270.2), or LINC01271 in cell viability assays. The mean cell numbers of three independent replicates of 4T1 control1, MaTAR25 KO1, MaTAR25 KO1 with GFP, LINC01270.1 clone1-5, LINC01270.2 clone1-5, and LINC01271 clone1-3 are shown ± SD (n = 3) *P < 0.05 (paired Student’s t test; two-tailed). c 4T1 MaTAR25 KO cells with ectopic expression of GFP was used as a control to assess rescue in a cell invasion assay. The mean relative cell invasion of two independent replicates of 4T1 control, MaTAR25 KO1, MaTAR25 KO1 with GFP clone1-2, LINC01271 ectopic expression clone1-2 is shown ± SD (n = 2 independent experiments). Ectopic expression of LINC01271 can rescue the MaTAR25 KO cell invasion phenotype. d RNA expression level of Tns1 was determined in MaTAR25 KO1 cells ectopically expressing LINC01270.1, LINC01270.2, or LINC01271 by qRT-PCR. Data are presented as mean values ± SD (n = 3 independent experiments). *P < 0.05 (paired Student’s t test; two-tailed). The protein level of mTns1 was also examined in MaTAR25 KO cells with ectopic expression of LINC01271 by immunoblot analysis. e Three different ASOs targeting LINC01271 were used to independently knockdown LINC01271 in MDA-MB-231 LM2 cells. Left panel: the knockdown efficiency is calculated and data are shown as mean values ± SD (n = 2 independent experiments) by qRT-PCR after 24 h treatment of ASOs. Right panel: the mean cell numbers of three independent cell counting experiments of MDA-MB-231 LM2 mock-treated control cells, cells treated with scrambled ASO, and cells treated with three different LINC01271 ASOs are shown ± SD (n = 3). *P < 0.05 (paired Student’s t test; two-tailed).

Fig. 7. LINC01271 expression in breast tumors and lung metastases.

a smRNA-FISH images showing the expression of LINC01271 (red) in patient breast tumor sections from different stages of breast cancer. More than six random areas on each slide were examined, and representative images are shown. Scale bars are 20 μm. The quantification of six images (three early stage and three late stage) of breast tumor sections is shown and data are presented as mean values ± SD. P = 0.08587 (paired Student’s t test; two-tailed). b smRNA-FISH images showing the expression pattern of LINC01271 (red) within luminal subtype patient breast cancer primary tumors and lung metastases sections from the same patients. More than six random areas on each slide were examined, and representative images are shown. Scale bars are 20 μm. c Proposed model of MaTAR25 function. MaTAR25 acts as a scaffold and/or chaperone recruiting PURB to the Tns1 gene where it induces its transcription. The Tns1 protein associates with focal adhesion complexes regulating signaling between the extracellular matrix and the actin cytoskeleton. KO or KD of MaTAR25 results in a reorganization of the actin cytoskeleton and focal adhesion complexes as well as a significant decrease in microvilli resulting in a reduction in cell proliferation and migration.