Identification of cis- and trans-acting factors involved in the localization of MALAT-1 noncoding RNA to nuclear speckles

Abstract

MALAT-1 noncoding RNA is localized to nuclear speckles despite its mRNA-like characteristics. Here, we report the identification of several key factors that promote the localization of MALAT-1 to nuclear speckles and also provide evidence that MALAT-1 is involved in the regulation of gene expression. Heterokaryon assays revealed that MALAT-1 does not shuttle between the nucleus and cytoplasm. RNAi-mediated repression of the nuclear speckle proteins, RNPS1, SRm160, or IBP160, which are well-known mRNA processing factors, resulted in the diffusion of MALAT-1 to the nucleoplasm. We demonstrated that MALAT-1 contains two distinct elements directing transcripts to nuclear speckles, which were also capable of binding to RNPS1 in vitro. Depletion of MALAT-1 represses the expression of several genes. Taken together, our results suggest that RNPS1, SRm160, and IBP160 contribute to the localization of MALAT-1 to nuclear speckles, where MALAT-1 could be involved in regulating gene expression.

FIGURE 1.

MALAT-1 is stably retained in the nucleus. The localization of human MALAT-1 RNA (green) and Flag-tagged hnRNP-A1/C1 (red) in HeLa-TO cells were examined by fluorescent in situ hybridization and immunocytochemistry, respectively. The nuclei of mouse NIH3T3 cells were easily distinguished from those of human HeLa cells (indicated by arrows) by the characteristic DAPI staining of intranuclear bodies (granules) in the mouse nuclei. Broken white lines in the merged images indicate the heterokaryon's boundaries.

FIGURE 2.

Subcellular localization of MALAT-1 fragments. (A) Variants of MALAT-1 RNA used in the experiments. Luc indicates luciferase cDNA. Images from the UCSC Genome Browser (http://genome.ucsc.edu/) represent the data of evolutionary conservations among 17 vertebrates. The human MALAT-1 locus is shown at the top with the graduated sizes. The results of the subcellular distribution analysis are summarized in the right column. C and N+C indicate distribution in the cytoplasm, and nucleoplasm with cytoplasm, respectively. (B) FISH analyses of HeLa-TO cells, transiently transfected with the indicated MALAT-1 variants (green) are displayed, together with immunocytochemistry staining with an anti-PCNA (blue) and anti-SC35 antibody (red). Both images were merged to represent the subcellular localization of the MALAT-1 derivatives. Scale bar, 5 μm.

FIGURE 2.

Subcellular localization of MALAT-1 fragments. (A) Variants of MALAT-1 RNA used in the experiments. Luc indicates luciferase cDNA. Images from the UCSC Genome Browser (http://genome.ucsc.edu/) represent the data of evolutionary conservations among 17 vertebrates. The human MALAT-1 locus is shown at the top with the graduated sizes. The results of the subcellular distribution analysis are summarized in the right column. C and N+C indicate distribution in the cytoplasm, and nucleoplasm with cytoplasm, respectively. (B) FISH analyses of HeLa-TO cells, transiently transfected with the indicated MALAT-1 variants (green) are displayed, together with immunocytochemistry staining with an anti-PCNA (blue) and anti-SC35 antibody (red). Both images were merged to represent the subcellular localization of the MALAT-1 derivatives. Scale bar, 5 μm.

FIGURE 3.

Subcellular localization of inner fragments derived from the M region of MALAT-1. (A) A map of inner fragments derived from the M region of MALAT-1. (B) FISH analysis of HeLa-TO cells transfected with the indicated inner MALAT-1 fragments (green), together with immunocytochemistry staining of PCNA (blue) and SC-35 (red). The merged images demonstrate the subcellular localization of the inner MALAT-1 fragments. (C) FISH analyses of HeLa-TO cells transfected with the MALAT-1 variants lacking either the E or M regions (ΔE, ΔM), or both (ΔEΔM) (green) are displayed, together with immunocytochemistry staining of PCNA (blue) and SC-35 (red). The merged images demonstrate that these deletion mutants do not localize in nuclear speckles.

FIGURE 4.

Effects of RNAi-mediated depletion of nuclear proteins on the localization of MALAT-1. (A) FISH analyses of HeLa-TO cells transfected with the indicated siRNAs (endogenous MALAT-1, green) are displayed, together with immunocytochemical staining of SRSF2 (SC35) (red) and SRm300 (cyan). DAPI (blue) was used to visualize nuclei. (B) Western blot analysis of the indicated proteins where expression had been knocked down. (C) Binding activity of RNPS1 to full-length or partial MALAT-1 RNA as measured by a filter-binding assay in vitro.

FIGURE 5.

Subcellular localization of U1 snRNA in cells where the expression of nuclear speckle proteins was repressed by RNAi. FISH analyses of HeLa-TO cells transfected with the indicated siRNAs (endogenous U1 snRNA, green) are displayed, together with immunocytochemical staining of SRSF2 (SC35) (red), SRm300 (cyan), and DAPI staining of DNA (blue).

FIGURE 6.

Subcellular distribution of poly(A)⁺ RNAs in cells where the expression of nuclear speckle proteins was repressed by RNAi. FISH analyses of HeLa-TO cells transfected with the indicated siRNAs (poly(A)⁺ RNA, green) are displayed, together with immunocytochemistry of SRSF2 (SC35) (red), SRm300 (cyan), and DAPI staining of DNA (blue).

FIGURE 7.

Reduced expression of OASL, IFI44, and SPINK4 in cells where the expression of MALAT-1, RNPS1, SRm160, or IBP160 was repressed by RNAi. (A) Total RNAs were prepared from cells transfected with siRNAs against MALAT-1 and then analyzed by qRT–PCR. Using two different siRNAs against MALAT-1 minimizes the possible off-target effects, causing apparent reductions of transcript levels. Each value was normalized to the expression level of glyceraldehyde 3-phosphate dehydrogenase (GAPDH). Error bars indicate the standard deviation (SD) obtained from three independent experiments. (B) Pre-mRNA levels of the indicated transcripts were measured by qRT–PCR. The primer set for OASL pre-mRNA and the primer set for IFI 44 pre-mRNA amplified the cDNA sequence in the exon 1 and intron 1 junction of each pre-mRNA. The primer set for SPINK4 pre-mRNA amplified the cDNA sequence in the junction between exon 3 and intron 3 of SPINK4. (C) The indicated transcripts were quantified by qRT–PCR analysis using total RNAs prepared from cells transfected with the indicated siRNAs. Each value was normalized to the expression level of GAPDH. Error bars represent the SD obtained from three independent experiments.