RNA gain-of-function in spinocerebellar ataxia type 8 - PubMed (original) (raw)

RNA gain-of-function in spinocerebellar ataxia type 8

Randy S Daughters et al. PLoS Genet. 2009 Aug.

Abstract

Microsatellite expansions cause a number of dominantly-inherited neurological diseases. Expansions in coding-regions cause protein gain-of-function effects, while non-coding expansions produce toxic RNAs that alter RNA splicing activities of MBNL and CELF proteins. Bi-directional expression of the spinocerebellar ataxia type 8 (SCA8) CTG CAG expansion produces CUG expansion RNAs (CUG(exp)) from the ATXN8OS gene and a nearly pure polyglutamine expansion protein encoded by ATXN8 CAG(exp) transcripts expressed in the opposite direction. Here, we present three lines of evidence that RNA gain-of-function plays a significant role in SCA8: 1) CUG(exp) transcripts accumulate as ribonuclear inclusions that co-localize with MBNL1 in selected neurons in the brain; 2) loss of Mbnl1 enhances motor deficits in SCA8 mice; 3) SCA8 CUG(exp) transcripts trigger splicing changes and increased expression of the CUGBP1-MBNL1 regulated CNS target, GABA-A transporter 4 (GAT4/Gabt4). In vivo optical imaging studies in SCA8 mice confirm that Gabt4 upregulation is associated with the predicted loss of GABAergic inhibition within the granular cell layer. These data demonstrate that CUG(exp) transcripts dysregulate MBNL/CELF regulated pathways in the brain and provide mechanistic insight into the CNS effects of other CUG(exp) disorders. Moreover, our demonstration that relatively short CUG(exp) transcripts cause RNA gain-of-function effects and the growing number of antisense transcripts recently reported in mammalian genomes suggest unrecognized toxic RNAs contribute to the pathophysiology of polyglutamine CAG CTG disorders.

PubMed Disclaimer

Conflict of interest statement

The authors have declared that no competing interests exist.

Figures

Figure 1

Figure 1. CUGexp–positive ribonuclear inclusions and co-localization with MBNL1 in SCA8 cerebellum.

(A) CUG-positive nuclear foci (RED) in SCA8 mouse and human cerebellum. CUG ribonuclear inclusions detected by FISH (RED) in SCA8 (human and mouse) and DM1 (human) Purkinje cells (PCs), molecular layer (ML) interneurons (human and mouse), and deep cerebellar nuclei (DCN) (mouse) showing CUG positive nuclear foci (RED) in human SCA8 and DM1 cerebellum (top) and in SCA8 BAC-EXP1 mice (bottom). (B) FISH and IF analysis of CUG foci and MBNL1 in human cerebellar autopsy tissue. CUG foci co-localize with MBNL1 in molecular layer interneurons (ML) (SCA8) (top row; arrowheads) but not in Purkinje cells (PC) (SCA8 or DM1) (middle and bottom rows, arrows). Co-localization is indicated by a yellowish color when the red CUG foci, green MBNL1 and blue DAPI overlap in the nucleus. (C) CUG foci co-localize with Mbnl1 in ML interneurons and deep cerebellar nuclei (DCN) in SCA8 BAC-EXP mice (arrowheads, top and bottom rows). Images are representative examples of staining with all 3 channels acquired at equal intensities in sequential order on a confocal microscope, merged along the Z-axis, digitally zoomed, cropped, and adjusted for brightness and contrast for publication.

Figure 2

Figure 2. Loss of MBNL1 enhances rotarod deficits in SCA8 BAC-EXP5 mice.

(A) Line plot of average latency to fall in seconds over four consecutive trial days. Individual trial day data points are the average of the last 3 trials and show a decreased latency for SCA8+/−; Mbnl1 +/ΔE3 mice that is significantly different from WT at day four. (B) Bar graph shows the mean latency to fall (Sec) for non-transgenic, singly transgenic SCA8 BAC-EXP5 (SCA8+/−). heterozygous Mbnl1 knock-out (Mbnl1+ /ΔE3) and doubly transgenic SCA8 BAC-EXP5; Mbnl1 +/ΔE3 (SCA8+/−; Mbnl1 +/ΔE3; n = 17) littermates. As previously reported for this low-copy line (SCA8 BAC-EXP5), no significant difference between non-transgenic littermates and singly transgenic SCA8+/− animals was found. Although singly transgenic Mbnl1+ /ΔE3 mice show a significant decrease in latency compared to non-transgenic (p = 0.01), doubly mutant SCA8+/−; Mbnl1+ /ΔE3 perform significantly worse than both WT and singly mutant Mbnl1+ /ΔE3 animals (p = 0.003). *and ‡ = significant differences between groups.

Figure 3

Figure 3. Alternative splicing changes in SCA8 and CUGBP1 RNA targets.

(A) RT-PCR of human autopsy tissue showing alternative splicing changes previously reported for DM1, NMDAR1 (exon 5) and MBNL1 (exon 7) also occur in SCA8. B) Diagram summarizing relative position of CUG-BP1 CLIP tags in genes. Boxes = exons (open = UTR; black-ORF); thick grey line = intergenic regions. Also indicated are tags that overlap intron/exon, orf/3′ UTR and 3′ UTR/intergenic junctions with the number of unique tags for each location in parentheses.

Figure 4

Figure 4. Upregulation of Gabt4 in SCA8 BAC-EXP1 and Mbnl1 ΔE3/ΔE3 mice.

(A) Example immunoblot showing increase in Gabt4 protein in SCA8 BAC-EXP1 and Mbnl1ΔE3/ΔE3 mice relative to strain specific WT controls. Bar graph shows average Gabt4 protein quantified by densitometry methods, and analyzed for mean between group differences compared to strain specific wild type animals. Asterisks = significant differences for SCA8 (p<0.002) and Mbnl1ΔE3/ΔE3 (p<0.001) mice compared to non-transgenic controls. (B) qRT-PCR showing increased Gabt4 vs. Hprt in SCA8 BAC-EXP and Mbnl1 ΔE3/ΔE3 compared to BAC-CTRL and non-transgenic littermates. Asterisks = statistical significance, p = 0.0015 for SCA8 and p<0.01 for Mbnl1 ΔE3/ΔE3 mice compared to non-transgenic controls. (C) Example protein blot showing no change in another prominent cerebellar GABA transporter receptor, Gabt1, in SCA8 BAC-EXP1 mice. (D) Fluorescent IHC staining showing a similar distribution with a qualitative increase in Gabt4 IF in cerebellar granular cell layer in SCA8 BAC-EXP1 compared to wildtype mice. (E) Optical responses in Crus II to stimulation of the ipsilateral C3 whisker pad in FVB and SCA8 mice showing increased cerebellar cortical response to whisker pad stimulation in SCA8 mice. Shown are the thresholded (mean±1 SD of the control region) and pseudocolored responses superimposed on an image of the background fluorescence. Scale bar = 1 mm. Both the intensity and area of the responses were greater in the SCA8 mice compared with FVB control mice. *denotes p<0.05.

Figure 5

Figure 5. Upregulation and alternative splicing of GABT4 in human SCA8 brain.

(A) qRT-PCR shows increased GABT4 RNA in frontal lobe autopsy tissue from adult SCA8 (n = 3) and control fetal (n = 1) tissue. * = significant difference (p<0.01). (B) Protein blot showing up-regulation of GABT4 in SCA8 and fetal frontal lobe vs. DM1 and Control with GAPDH as loading control. (C) RT-PCR of GABT4 shows a shift favoring exon 7 inclusion transcripts in SCA8 and control fetal frontal lobe tissue vs. adult control and DM1.

Figure 6

Figure 6. Endogenous GABT4 up-regulation in SK-N-SH cells induced by CUGexp but not CAGexp transcripts.

(A) Schematic of constructs used to express SCA8 CUGexp or CAGexp and pure CUGexp and CAGexp transcripts. (B) qRT-PCR shows increased GABT4 RNA levels in SK-N-SH cells expressing Exon A CUGexp or pure polyCUGexp transcripts. * = statistical significance, p<0.0001. (C) Alternative splicing shifts favoring exon 7 inclusion in cells expressing SCA8 Exon A CUGexp or pure polyCUGexp that does not occur in response to SCA8 Exon A CAG or pure polyCAGexp transcripts compared to vector alone.

Figure 7

Figure 7. Antagonistic Regulation of GABT4 by CUGBP1 and MBNL1 in SK-N-SH cells.

(A,B) qRT-PCR showing increased GABT4 transcript levels and alternative splicing changes favoring exon 7 inclusion in cells overexpressing CUGexp transcripts or CUG-BP1 but not in cells overexpressing MBNL1 1/41 or CUGexp and MBNL1 1/41. (C) Protein blot showing GABT4 is increased in cells overexpressing CUGexp transcripts or CUG-BP1, but not in cells overexpressing MBNL1/41 alone or MBNL1/41 and CUGexp transcripts.

Comment in

Similar articles

Cited by

References

    1. Day JW, Schut LJ, Moseley ML, Durand AC, Ranum LPW. Spinocerebellar ataxia type 8: clinical features in a large family. Neurology. 2000;55:649–657. - PubMed
    1. Koob MD, Moseley ML, Schut LJ, Benzow KA, Bird TD, et al. An untranslated CTG expansion causes a novel form of spinocerebellar ataxia (SCA8). Nat Genet. 1999;21:379–384. - PubMed
    1. Moseley ML, Zu T, Ikeda Y, Gao W, Mosemiller AK, et al. Bidirectional expression of CUG and CAG expansion transcripts and intranuclear polyglutamine inclusions in spinocerebellar ataxia type 8. Nat Genet. 2006;38:758–769. - PubMed
    1. He Y, Zu T, Benzow KA, Orr HT, Clark HB, et al. Targeted deletion of a single Sca8 ataxia locus allele in mice causes abnormal gait, progressive loss of motor coordination, and Purkinje cell dendritic deficits. J Neurosci. 2006;26:9975–9982. - PMC - PubMed
    1. Ranum LPW, Cooper TA. RNA-mediated neuromuscular disorders. Annu Rev Neurosci. 2006;29:259–277. - PubMed

Publication types

MeSH terms

Substances

Grants and funding

LinkOut - more resources