Fig 2: NEAT1 interactors do not dissociate upon DNA damage–induced PS formation.(A) Top: RNA FISH for total NEAT1 (red) NEAT1_2 (green) nucleus (blue). Bottom: RNA FISH combined to IF for total NEAT1 (red), NEAT1_2 (magenta), PSPC1 (green), and nucleus (blue) in MCF-7 exposed to HU. (B) RT-qPCR of total NEAT1 transcript (NEAT1) and long form (NEAT1_2), in MCF-7 exposed to HU after a pulse of actinomycin D (ActD). SRSF1 is a positive control, and 16S is a negative control. Error bars represent means ± SD. P values were calculated by two-way ANOVA, using three biological replicates. *P < 0.05, ***P < 0.001, and ****P < 0.0001. (C) RAP-MS experimental approach in MCF-7 treated with HU. In the volcano plots, significantly enriched protein candidates (t test, P < 0.05 and FC > 1.6) are color-coded by protein complexes. (D) RAP Western blot in MCF-7 for PS proteins (positive controls) and the newly identified NEAT1 interactors. H3 is a negative control. Input (INP) loaded is either 1 or 2% of the total nuclear lysate. (E) STORM images of the PS protein NONO (red) and INTS1 (green) in MCF-7 treated with HU. Voronoi diagram and quantification of NONO (class A) and INTS1 (class B) colocalization are shown.

Fig 3: Integrator restrains PS biogenesis.(A) RNA-seq (total and small) of HeLa shINTS11 or shCtrl at NEAT1 locus. The blue shade indicates NEAT1 termination site. Dox, doxycycline. (B) RNA-seq (total and small) of HeLa expressing INTS11 catalytic dead mutant (E203Q), empty vector (empty), or WT INTS11 at NEAT1 locus. (C) RT-qPCR for NEAT1 and NEAT1_2 expression (relative to siC) in MCF-7 upon siINTS11. RNU11 and RNU12 and their read-through are positive controls. Error bars represent means ± SD. P values were calculated by two-tailed t test, using three biological replicates. ***P < 0.001 and ****P < 0.0001. (D) FISH quantification for NEAT1 and NEAT1_2. Error bars represent means ± SD. P values were calculated by paired two-tailed t test. ****P < 0.0001. (E) Immunofluorescence (IF) for the PS protein PSPC1 (green) and RNA FISH for NEAT1 in MCF-7 cells siINTS11 (total NEAT1, red; NEAT1_2, magenta; nuclei, blue). Right: 3D quantification peaks represent signal intensities at colocalization site.

Fig 4: NEAT1 interactome is enriched in 3' end processing factors.(A) RAP–mass spectrometry (MS) strategy to study NEAT1 interactome. Probes targeting either the melanoma-specific LINC00698 (Ctrl) or the 5' end of NEAT1 short and long isoforms (N1_5') were used. (B) Relative abundance by RT-qPCR of the S14, NEAT1, and MALAT1 in MCF-7 cytoplasmic and nuclear fractions. (C) Both form-specific (NEAT1) and long form-specific (NEAT1_2) RT-qPCR to evaluate the efficiency of the RAP. Three abundant coding and noncoding transcripts (MALAT1, TBP, and HPRT1) are shown as negative controls. Error bars represent means ± SD. P values were calculated by two-way analysis of variance (ANOVA), using three biological replicates. ****P < 0.0001. (D) Venn diagram showing overlap between NEAT1 previously identified partners (orange) and candidates identified in this study (brown). In purple, the interactors considered for further analysis. (E) Volcano plots indicating the interactors significantly enriched (t test, P < 0.05 and FC > 1.6) by N1_5' probes. Highlighted candidates are color-coded by protein complexes. The x axis indicates the ratio N1_5'/Ctrl in log2 scale. The y axis is the –log10 (t test, P value). (F) RAP Western blot validation of novel interactors. PS proteins PSF, PSPC1, NONO, and TDP-43 were used as positive controls, and H3 was used as negative control. Input (INP) was either 1 or 2% of the total nuclear lysate.

Fig 5: Paraspeckle protein NONO presents aggregated clusters in hematopoietic cells’ nuclei and reduced expression in granulocytes(A) Confocal microscopy analysis of NONO, a paraspeckle protein, using an immunofluorescent staining technique, in representative primary cell types through hematopoiesis, HSC, MPP, CMP, GMP, MEP, granulocyte, B cell. NONO (red) is superimposed over nuclei stained with DAPI (blue). Scale bar denotes 10 µ.m. (B and C) Expression levels of mRNA of the paraspeckles proteins NONO and PSPC1, respectively. Data is from ImmGen’s microarray database, showing averages of at least triplicates per cell type except for MPP and MEP that are duplicates; and SD per cell type. (D) Quantification of the foci number per cell type from immunofluorescent stained cells for NONO. Histograms show averages ±SD.