Altered Levels of Long NcRNAs Meg3 and Neat1 in Cell And Animal Models Of Huntington's Disease

Abstract

Altered expression levels of protein-coding genes and microRNAs have been implicated in the pathogenesis of Huntington's disease (HD). The involvement of other ncRNAs, especially long ncRNAs (lncRNA), is being realized recently and the related knowledge is still rudimentary. Using small RNA sequencing and PCR arrays we observed perturbations in the levels of 12 ncRNAs in HD mouse brain, eight of which had human homologs. Of these, Meg3, Neat1, and Xist showed a consistent and significant increase in HD cell and animal models. Transient knock-down of Meg3 and Neat1 in cell models of HD led to a significant decrease of aggregates formed by mutant huntingtin and downregulation of the endogenous Tp53 expression. Understanding Meg3 and Neat1 functions in the context of HD pathogenesis is likely to open up new strategies to control the disease.

Keywords: Huntington’s disease; MEG3; NEAT1; long ncRNA.

Figure 1.

Heatmap of statistically significant differentially regulated ncRNAs Xist(NR_001463), Meg3(NR_003633), Peg3os(NR_023846), Vaultrc5(NR_027885),Snora21 (NR_028078),Gm12238(NR_028480),Snord53(NR_028551),Snord85(NR_028565),Snhg12(NR_029468),Snord42a(NR_037682),Gm22650(NR_128564),Gm38671(NR_128567),Neat1(NR_131212) in 6 weeks (early stage) and 8 weeks (late stage) old Huntington’s mouse cortex compared to the control cortex from small RNA sequencing. Each sample has at least two biological replicates. Color codes indicate normalized fold changes- Red = upregulation; Green = downregulation.

Figure 2.

Altered levels of ncRNAs in HD animal model and in cells expressing N-terminal mutant Huntingtin. (i). Bar graphs representative of three (n = 3) independent experiments measuring levels of Brip1os, Meg3, Neat1, Snhg3and Snhg12 (panels A and C); Snora21, Snord53, Snord85, Vaultrc5 and Xist (panels B and D) by qRT-PCR in cortex region of 6-weeks old (panels A and B) and 8-weeks old (panel C and D) R6/2 mice and age-matched wild-type mice. (ii).Bar graphs representative of three (n = 3) independent experiments measuring levels of Brip1os, Meg3, Neat1, Snhg3and Snhg12 (panel A); Snora21, Snord53, Snord85, Vaultrc5 and Xist (panel B) by qRT-PCR in mouse immortalized striatal cells expressing full-length huntingtin (Hdh) gene with 7 (STHdh^Q7/Hdh^Q7 cells) and 111 (STHdh^Q111/Hdh^Q111cells) glutamine repeats.(iii) Bar graphs representative of three (n = 3) independent experiments measuring levels of Brip1os, Meg3, Neat1, Snhg3and Snhg12 (panel A); Snora21, Snord53, Snord85, Vaultrc5 and Xist (panel B) by qRT-PCR in STHdh^Q7/Hdh^Q7 cells transiently expressing empty DsRed vector or huntingtin exon 1 having 16 and 83 glutamine repeats cloned in DsRed vector (designated as 16Q-DsRed and 83Q-DsRed respectively). Levels of β-actin were taken as endogenous control. The levels of individual ncRNAs were normalized by the corresponding β-actin levels. Fold change was calculated by considering the relative levels of ncRNA in empty vector (DsRed) transfected cells (control) to be 1.(iv) Bar graphs representative of three (n = 3) independent experiments measuring levels of Meg3, Neat1, and Xist) by qRT-PCR in Neuro2A cells transiently expressing huntingtin exon 1 having 16 and 83 glutamine repeats cloned in DsRed vector (designated as 16Q-DsRed and 83Q-DsRed respectively). Levels of β-actin were taken as endogenous control. The levels of individual ncRNA were normalized by the corresponding β-actin levels. Error bars indicate ± SD. The statistical significance level between different experimental pairs is indicated (NS, not significant; *, p < 0.05; **, p < 0.01; ***, p < 0.001).

Figure 2.

Continued.

Figure 3.

Reduction in the numbers of HTT-83Q-DsRed aggregates in Neuro2A and SHSY5Y cells. (i) & (iv) HTT-83Q-DsRed aggregates in Neuro2A and SHSY5Y cells respectively, (ii) & (v) HTT-83Q-DsRed aggregates in Neuro2A and SHSY5Y cells respectively co-transfected with siRNA against Neat1 (iii) & (vi) HTT-83Q-DsRed aggregates in Neuro2A and SHSY5Y cells respectively co-transfected with siRNA against Meg3. All representative images were acquired 24 hours post-transfection.(vii) Bar graphs representative of three (n = 3) independent experiments, taking 30 cells each time shows the decrease in aggregate numbers per cell in cells transfected with HTT-83Q-DsRed and treated with siRNAs against Meg3 or Neat1 compared to cells transfected with HTT-83Q-DsRed only. Error bars represent standard deviation, *represents statistical significance (*p ≤ 0.05; **p ≤ 0.01).

Figure 3.

Continued.

Figure 4.

Endogenous p53 protein levels alter after treatment with siRNA against Meg3 or Neat1 in HD cell model. (i) Representative western blot of three independent experiments (n = 3) shows (Left to Right) decrease in endogenous levels of p53 in wild-type (WT) Neuro2A cells, Neuro2A cells co-transfected with HTT-83Q-DsRed and siRNA against Meg3 or Neat1, Neuro2A cells transfected with HTT-83Q-DsRed only and Neuro2A cells transfected only with siRNA against Meg3 or Neat1 without HTT-83Q-DsRed, 24 hours post-transfection.(ii) Histogram representing the fold changes of p53 in the different conditions compared to WT cells in (i) normalized to β-actin. Fold change was calculated by considering the relative levels of p53 in WT cells (control) to be 1. Error bars represent standard deviation, *represents statistical significance (*p ≤ 0.05; **p ≤ 0.01).

Figure 5.

Comparison of Biological processes in noncoding RNA interacting proteins (dark black) and in human Genome (light black). GO ID is shown within parenthesis, except for Regulation of transcription (GO: 0006355) and Nuclear mRNA splicing (GO: 0000398). Levels of significance after multiple testing between the two as determined in the GeneCodis3 analysis are: 1.39E⁻²⁷, 3.67E⁻¹⁶, 1.52E⁻¹⁵, 2.89E⁻¹⁵, 1.69E⁻¹³, 1.83E⁻¹³, 1.69E⁻⁰⁹, 1.69E⁻⁰⁹, 1.55E⁻⁰⁷, 1.13E⁻⁰⁶, 2.15E⁻⁰⁶ and 2.72E⁻⁰⁶ respectively from left to right.