The levels of the long noncoding RNA MALAT1 affect cell viability and modulate TDP-43 binding to mRNA in the nucleus

Abstract

TAR DNA-binding protein (TDP-43) and metastasis-associated lung adenocarcinoma transcript (MALAT1) RNA are both abundantly expressed in the human cell nucleus. Increased interaction of TDP-43 and MALAT1, as well as dysregulation of TDP-43 function, was previously identified in brain samples from patients with neurodegenerative disease compared to healthy brain tissues. We hypothesized that TDP-43 function may depend in part on MALAT1 expression levels. Here, we find that alterations in MALAT1 expression affect cell viability and can modulate TDP-43 binding to other mRNAs in HEK293 and SH-SY5Y human cell lines. Disruption of either MALAT1 or TDP-43 expression induces cell death, indicating that both macromolecules contribute positively to survival. Depletion of MALAT1 RNA results in increased binding of TDP-43 to other mRNA transcripts at the 3' UTR. Finally, we examined the contribution of MALAT1 expression to survival in a cell culture model of neurodegeneration using MPP⁺ treatment in SH-SY5Y cells. Depletion of MALAT1 RNA protects against toxicity in a cellular model of neurodegeneration and modulates TDP-43 binding to mRNA transcripts involved in apoptotic cell death. Taken together, we find that MALAT1 RNA and TDP-43 interactions can affect mRNA levels and cell viability. A tightly regulated network of noncoding RNA, messenger RNA, and protein interactions could provide a mechanism to maintain appropriate RNA expression levels and contribute to neuronal function.

Keywords: Long non-coding RNA; MALAT1; RNA binding protein; RNA-protein interaction; TAR DNA-binding protein 43 (TDP-43) (TARDBP); gene expression; mRNA; mRNA stability; neurodegeneration.

Figure 1

Disruption of MALAT1 or TDP-43 induces cell death and perturbs the RNA transcript levels of the other binding partner in SH-SY5Y neuroblastoma cells.A, cell viability after overexpression or knockdown of TDP-43. N = 3 biological replicates. B, RNA expression changes in TARDBP mRNA and MALAT1 RNA levels, after TARDBP/TDP-43 overexpression or knockdown. C, cell viability after overexpression or knockdown of MALAT1 noncoding RNA. N = 6 biological replicates. D, gene expression changes in TARDBP mRNA and MALAT1 RNA levels, after MALAT1 overexpression or knockdown. E, protein expression changes in TDP-43 after MALAT1 overexpression. F, protein expression changes in TDP-43 after MALAT1 knockdown. N = 3 for all gene expression and Western blot experiments. T tests are conducted with two tailed unpaired equal variance conditions. ∗ = p < 0.05, ∗∗ = p < 0.01. All data are plotted with SD.

Figure 2

MALAT1 knockdown induces cell death in SH-SY5Y cells but does not lead to cytoplasmic localization of TDP-43.A, immunofluorescence imaging of actin with DAPI nuclear staining, in SH-SY5Y cells after 48 h of transfection with negative control (NC) GapmeR or MALAT1 GapmeR. Scale bar represents 30 μm. B, quantification of nuclear area in NC GapmeR– or MALAT1 GapmeR–treated cells. N = 51 for NC GapmeR, N = 91 for MALAT1 GapmeR. C, classification of apoptotic and aberrant nuclei in NC GapmeR– or MALAT1 GapmeR–treated cells. N = 51 for NC GapmeR, N = 91 for MALAT1 GapmeR. D, immunofluorescence imaging of TDP-43 with DAPI nuclear staining in SH-SY5Y cells after 48 h of transfection with NC or MALAT1 GapmeR. Scale bars represent 30 μm. E, Western blot analysis of TDP-43 protein level in cytoplasmic and nuclear fractions purified from HEK293 cells transfected for 24 h with NC or MALAT1 GapmeR. F, quantification of TDP-43 protein distribution percentage across nuclear and cytoplasmic fractions from Western blot data, N = 3 biological replicates. T tests are conducted with two tailed unpaired equal variance conditions. ∗ = p < 0.05, ∗∗ = p < 0.01. All data are plotted with SD.

Figure 3

TDP-43 binds the 3′ UTR of common RNA targets in multiple cell types.A, overlap of top 5000 transcripts bound by TDP-43 from CLIP-seq datasets from SH-SY5Y neuroblastoma cells (ERR039855), H9 human embryonic stem cells (SRR4044755), and HEK293 cells (ERR9192743). B, gene ontology analysis of the 1549 commonly bound genes among the three CLIP-seq datasets. C, log fold enrichment of TDP-43–binding regions in the commonly bound transcript set. D, Integrative Genomics Viewer CLIP-seq read profiles of TDP-43 binding to HMGB2, SLC1A5, and CSNK1E mRNA transcripts, the negative control BARD1 mRNA transcript, and MALAT1 RNA transcript regions. Reads were visualized from the SH-SY5Y TDP-43 CLIP-Seq dataset ERR039855.

Figure 4

TDP-43 binds the 3′ UTR of messenger RNAs with high affinity in vitro. A, representative image of multiplexed EMSA for TDP-43 binding to 3′ UTR fragment of CSNK1E mRNA, positive control TARDBP mRNA region, and negative control BARD1 mRNA region. B, Hill’s plot for TDP-43 binding with CSNK1E, TARDBP, and BARD1 RNA transcripts. At least three experimental replicate assays for TDP-43 binding to each RNA fragment were performed, and data are plotted with SD. C, representative image of multiplexed EMSA for TDP-43 binding to mRNA 3′ UTR fragments from HMGB2, SLC1A5, and CSNK1E. D, Hill’s plot for TDP-43 binding with 3′ UTR fragments from HMGB2, SLC1A5, and CSNK1E transcripts. At least three experimental replicate assays for TDP-43 binding to each RNA fragment were performed, and data are plotted with SD. E, competitive EMSA for TDP-43 RRMs bound to Cy5-labeled TARDBP RNA with addition of excess unlabeled TARDBP RNA. F, competitive mEMSA for TDP-43 RRMs bound to Cy5-labeled TARDBP positive control with addition of excess unlabeled BARD1 negative control RNA. G, quantitation of TARDBP RNA fraction bound in competition assays for each sample pair. N = 2 biological replicates for BARD1 negative control. N = 3 biological replicates for TARDBP positive control. Data are plotted with SD.

Figure 5

Reduced level of MALAT1 RNA results in relocalization of TDP-43 binding to messenger RNA transcripts in HEK293 cells.A, IP-qPCR analysis of TDP-43 enrichment of HMGB2, SLC1A5, CSNK1E, and BARD1 mRNA transcripts, MALAT1 RNA, and TARDBP mRNA control compared to IgG enrichment of each transcript. B, representative Western blot of TDP-43 protein recovery after immunoprecipitation with TDP-43 or IgG control antibody, from negative control (NC) GapmeR– or MALAT1 GapmeR–treated cells. C, IP-qPCR analysis of RNA quantification for TDP-43 enrichment of each RNA transcript after knockdown of MALAT1 RNA, compared to TDP-43 enrichment of each RNA transcript in NC GapmeR–treated cells. D, gene expression changes in 3′ UTR target genes after knockdown of MALAT1 RNA or TDP-43 protein, compared to negative control (NC) GapmeR treatment. E, Western blot for CSNK1E, SLC1A5, and HMGB2 protein levels after 48 h of transfection with MALAT1 GapmeR or NC GapmeR. F, quantification of protein levels from Western blots in (E). All experiments in this figure were performed in HEK293 cells with N = 3 biological replicates. T tests are conducted with two tailed unpaired equal variance conditions. ∗ = p < 0.05, ∗∗ = p < 0.01, ns = p > 0.05. All data are plotted with SD.

Figure 6

Depletion of MALAT1 RNA protects against MPP + toxicity in a cell culture model of neurodegeneration.A, quantification of MALAT1 RNA levels in WT- or MPP⁺-treated SH-SY5Y cells. B, quantification of MALAT1 RNA levels in MPP⁺ cells treated with either negative control (NC) GapmeR or MALAT1 GapmeR. C, relative expression of mRNA transcripts HMGB2, SLC1A5, and CSNK1E after MPP+ treatment, compared to expression levels in untreated (WT) SH-SY5Y cells. D, relative expression of mRNA transcripts HMGB2, SLC1A5, and CSNK1E in MPP⁺ condition after treatment with negative control (NC) GapmeR or MALAT1 GapmeR. E, IP-qPCR for TDP-43 binding to MALAT1 RNA, HMGB2, CSNK1E, SLC1A5, and BARD1 mRNA in MPP⁺-treated cells compared to untreated (WT) SH-SY5Y cells. N = 2. F, viability of SH-SY5Y cells after MPP⁺ treatment compared to untreated (WT) cells. G, viability of SH-SY5Y cells after treatment with the indicated combinations of MPP⁺, TDP-43 GapmeR, MALAT1 GapmeR, pcEV empty vector control, pcMALAT1 overexpression construct, or pcTDP-43 overexpression construct. All experiments in this figure were performed in SH-SY5Y cells. N = 3 for RNA expression experiments, N = 3 for IP experiments, and N = 6 for cell viability experiments. T tests are conducted with two tailed unpaired equal variance conditions. ∗ = p < 0.05, ∗∗ = p < 0.01 All data are plotted with SD.

Figure 7

Proposed model for MALAT1 modulation ofTDP-43binding to 3′ UTR of messenger RNAs.