Interrogation of brain miRNA and mRNA expression profiles reveals a molecular regulatory network that is perturbed by mutant huntingtin

Abstract

Emerging evidence indicates that microRNAs (miRNAs) may play an important role in the pathogenesis of Huntington's disease (HD). To identify the individual miRNAs that are altered in HD and may therefore regulate a gene network underlying mutant huntingtin-induced neuronal dysfunction in HD, we performed miRNA array analysis combined with mRNA profiling in the cerebral cortex from N171-82Q HD mice. Expression profiles of miRNAs as well as mRNAs in HD mouse cerebral cortex were analyzed and confirmed at different stages of disease progression; the most significant changes of miRNAs in the cerebral cortex were also detected in the striatum of HD mice. Our results revealed a significant alteration of miR-200 family members, miR-200a, and miR-200c in the cerebral cortex and the striatum, at the early stage of disease progression in N171-82Q HD mice. We used a coordinated approach to integrate miRNA and mRNA profiling, and applied bioinformatics to predict a target gene network potentially regulated by these significantly altered miRNAs that might be involved in HD disease progression. Interestingly, miR-200a and miR-200c are predicted to target genes regulating synaptic function, neurodevelopment, and neuronal survival. Our results suggest that altered expression of miR-200a and miR-200c may interrupt the production of proteins involved in neuronal plasticity and survival, and further investigation of the involvement of perturbed miRNA expression in HD pathogenesis is warranted, and may lead to reveal novel approaches for HD therapy.

Fig. 1

Schematic flow diagram describing procedures used to identify inversely correlated putative target genes of the differentially expressed miRNAs in HD mice.

Fig. 2. Significantly altered miRNAs in the cerebral cortex of N171-82Q HD mice by miRNA microarray analysis

From 382 distinct miRNAs included on our array, (a) 8 miRNAs were significantly upregulated and 7 miRNAs were downregulated in cerebral cortex of 12-week-old HD mice compared to those in their age-matched littermate nontransgenic controls. (b) 16 miRNAs were upregulated and 8 miRNAs were downregulated in cerebral cortex of 18-week-old HD mice compared to those in their age-matched littermate controls. Blue bars represent previously described brain-enriched miRNAs, and red bars represent newly identified miRNAs in brains in our study. (c) miRNAs were significantly upregulated in the cerebral cortex of 12-week-old HD mice compared to those in 18-week-old HD mice. The criteria for the selection of significantly changed miRNAs is fold change >1.2, FDR<0.3, and p<0.05 (Z-test). N=3 samples each group.

Fig 3. Significantly altered gene expression and Z-ratio values were identified between HD mice and wild type controls by microarray analysis

Top panel: Venn diagram summarizes distinct and overlapping expression of mRNAs in cerebral cortex of HD mice compared to their controls at 12 and 18 weeks of age. Numbers of upregulated gene (Red) and downregulated gene (Blue). Note that there are 71 upregulated genes and 149 downregulated genes present only in 12-week-old HD mouse brains; and 155 genes are upregulated, and 79 genes are downregulated only in 18-week-old HD mice. Twelve upregulated genes and 27 downregulated genes are commonly present in both 12-week-old and 18-week-old HD mouse brains. Bottom colored bar graphs represent Z-ratios of top 100 significantly altered mRNAs. Criteria for the selection of significantly changed mRNAs are: fold change >1.2, FDR<0.3, and p<0.05 (Z-test). n=3

Fig. 4. Significantly altered gene expression and Z-ratio were identified in cerebral cortex of 12-week-old mice and 18-week-old mice by microarray analysis

Top panel: Venn diagram summarizes distinct mRNA expression profiles and overlapping genes in HD mice along with disease progression or wild type (WT) control mice along with age. Numbers represent upregulated gene (Red), and downregulated gene (Blue). Note that there are 41 genes that are upregulated and 40 genes are downregulated in 18-week-old HD mouse brains compared to those in 12-week-old HD mouse brains, while 98 genes are upregulated and 121 genes are downregulated in 18-week-old wild type mouse brains compared to those in 12-week-old wild type mouse brains. Only 2 genes that are upregulated and 3 genes that are downregulated commonly exist in both HD and wild type mouse brains as age increases from 12 weeks to 18 weeks. Bottom Z-ratio graphs show top 100 significantly altered genes between 18-week-old HD mouse brains and 12-week-old HD brains (left) or between 18-week-old wild type mouse brains versus 12-week-old wild type mouse brains, according to the criteria for the selection of significantly changed mRNAs, i.e, fold change >1.2, FDR<0.3, and p<0.05 (Z-test). n=3.

Fig. 5. Significantly altered miR-200a and miR-200c were detected in the cortex and striatum of N171-82Q HD mice by quantitative RT-PCR

MiR-200a levels were significantly upregulated in the cortex (a) and striatum (b) of 12-week-old HD mice; miR-200c levels were significantly upregulated in the cortex (c) and striatum (d) of 12-week-old HD mice. Bars represent Mean and STD, n= 5–6 mice in each group. *p<0.05 vs the levels of age-matched littermate controls by Student’s t-tests.

Fig. 6. Dynamic changes of miR-200a and miR-200c from presymptomatic stage to late disease progression in the cortex and striatum of N171-82Q

(a) Levels of miR-200a and miR-200c in the cortex of N171-82Q at different ages. (b) Levels of miR-200a and miR-200c in the striatum of N171-82Q mice at different ages. Bars represent Mean and STD, n= 5–6 mice in each group. *p<0.05 vs the levels of age-matched littermate controls by Student’s t-tests.

Fig. 7. Trim2 is a target of miR-200a

(a) Immortalized striatal cells expressing mutant HTT (STHdh^Q111/Q111 cells) were transfected with pPRIME-CMV-GFP-FF3-pre miR-200a. Cells with green fluorescence represent transfected cells. (b) STHdh^Q111/Q111 cells transfected with pre-miR-200a produce mature miR-200a detected by real-time qPCR. (c) Quantification of miR-200a levels in STHdh^Q111/Q111 cells transfected with pre-miR-200a. (d) DNA sequencing confirmed the mutation of two binding sites of mouse Trim 2 full length 3′UTR: from ‘agtgtt’ to ‘CgCgtC’ and from ‘cagtgttt’ to ‘GCgCgtCt’, respectively, by site mutagenesis. (e) Luciferase activity assayed in STHdh^Q111/Q111 cells transfected with miR-200a and 3′UTR regions of Trim2 or 3′-UTR with mutation of miR-200a binding sites (Trim2 3′UTR-mut). Mean ± STD, n= 3. *p<0.05 vs Trim2 3′UTR transfected group; **p<0.05 vs Trim2 3′UTR+miR-200a transfected group by ANOVA followed by Scheffe’s post-hoc analysis. (f) Western blots of Trim2 levels in striatal cells transfected with miR-200a or antisense to miR-200a. Cells were collected 24 h after transfection.