EXOSC8 mutations alter mRNA metabolism and cause hypomyelination with spinal muscular atrophy and cerebellar hypoplasia

Abstract

The exosome is a multi-protein complex, required for the degradation of AU-rich element (ARE) containing messenger RNAs (mRNAs). EXOSC8 is an essential protein of the exosome core, as its depletion causes a severe growth defect in yeast. Here we show that homozygous missense mutations in EXOSC8 cause progressive and lethal neurological disease in 22 infants from three independent pedigrees. Affected individuals have cerebellar and corpus callosum hypoplasia, abnormal myelination of the central nervous system or spinal motor neuron disease. Experimental downregulation of EXOSC8 in human oligodendroglia cells and in zebrafish induce a specific increase in ARE mRNAs encoding myelin proteins, showing that the imbalanced supply of myelin proteins causes the disruption of myelin, and explaining the clinical presentation. These findings show the central role of the exosomal pathway in neurodegenerative disease.

Figure 1. Pedigrees with clinical presentation and brain MRI.

(a) Pedigree of the original Hungarian Roma family. *DNA of these family members was used for mutation analysis. (b) Pedigree 2, Hungarian Roma ethnic origin. (c) Patient P-II:10 at age 6 months. (d) MRI of patient P-II:10-detected immature myelination, which was consistent with the patient’s age (5 months, axial T2 image) A moderately thin corpus callosum was seen in the sagittal T2 image of the same patient. The extra-cerebral cerebrospinal fluid spaces were satisfactory in appearance. (e) Pedigree 3, a consanguineous Palestinian family. (f) Brain MRI of the affected siblings revealed mega cisterna magna and hypoplasia of the cerebellum and corpus callosum.

Figure 2. Autopsy findings.

(a–h) Cerebral cortex and white matter in autopsy of patient V:3 (pedigree 1). (a,b) Thickness of the white matter is significantly reduced, while the cortex is relatively well preserved. Black arrow: thickness of the cortex from the pia to the cortex/white matter borderzone; grey arrow: thickness of the white matter from the borderzone to the ependymal lining (Hematoxylin and eosin (H&E) staining); extensive astrogliosis (pale nuclei; white arrowhead) and reduced frequency of oligodendrocytes (dark round nuclei; black arrowhead). (c) Cerebellar cortex is relatively well preserved, but there is a microvacuolation of the underlying white matter within the cerebellar folia (H&E staining). (d) Reactive astrocytes (arrows) within the cerebellar white matter (vimentin immunohistochemistry). (e,f) Purkinje cell axonal torpedos (modified Bielschowsky silver method) (e) and neurofilament antibodies (f) indicating the loss of axonal connections. (g) Loss of myelin within the cerebellar white matter (Luxol Fast Blue). (h) Relative preservation of the myelin staining on myelin basic protein immunohistochemistry. (i–n) Spinal cord: normal control (i,l), Pelizaeus–Merzbacher disease (j,m) and EXOSC8 deficiency (patient V:20) (k,n). In EXOSC8 deficiency, myelin basic protein is present–apart from the longitudinal descending fibre tracts (k, *)–while it is completely absent in Pelizaeus–Merzbacher disease (j). Myelin is well preserved within the peripheral nerve roots (m,n, arrowheads), while indicates severe loss of myelin in both conditions within the spinal cord (m,n, Luxol Fast Blue).

Figure 3. Homozygosity mapping, exome sequencing, EXOSC8 mutation analysis and immunoblotting for EXOSC8.

(a) Genome-Wide Human SNP Array 6.0 (Affymetrix) was performed in six affected (P1-V:2, P1-V:10, P1-V:20, P1-V:27, P1-V:29, P1-VI:3) and one unaffected family members (P1-IV:14). Homozygosity mapping indicated two regions of homozygosity on chromosome 13 spanning from 36214563, bp to 37767059, bp (rs7327540 to rs11147637) and 43243791, bp to 44640995, bp (rs9533208 to rs9567354). (b) Exome sequencing detected 2,294 and 2,274 rare/novel single nucleotide variants (SNV) of which 184 were shared homozygous variants in patients P1-V:10 and P1-V:29; seven were protein altering, one of which was in the region of homozygosity in EXOSC8. SNV-Varscan parameters (*) minimum total coverage≥fivefold, minimum variant coverage ≥onefold, minimum quality>10; Indel-Dindel output filter minimum variant coverage ≥4. (a) Variants with position within targets (Illumina Truseq 62 Mb) ±500 bp, seen on both (forward and reverse) strands and (SBVs only) variant allele frequency >24%. (b) Variants that match 1,000 Genomes (Feb2012) and/or NHLBI 6500 exomes and/or In-house 334 exomes with MAF>0.01. (c) Confirmatory Sanger Sequencing indicated the presence of the homozygous c.815 G>C, p.Ser272Thr mutation in an affected patient (A) and heterozygously in her mother (U). This rare variant is listed as rs36027220 with a minor allele frequency of <0.01 (ref database) and was present also in pedigree 2. In pedigree 3 another homozygous mutation c.5C>T, p.Ala2Val has been identified (control - upper chromatogram, patient - lower chromatogram). (d) Immunoblotting showing EXOSC8 protein in control myoblasts (C. M.) and in myoblasts of P1-V:10 (P. M.); control fibroblasts (C. F.) and in fibroblasts of P3-II:1 (P. F.) using β-Actin or GAPDH as a loading control. (e) Both mutations alter conserved amino acids. (f) Crystal structure of the human exosome complex. The cap proteins are indicated in green (EXOSC3 light green) and the core proteins in grey. EXOSC8 highlighted in pink and the mutated residues (Ser 272, Ala2) are marked as black (arrowhead).

Figure 4. Characterization of the yeast RRP43 and tetRRP43 strains to access mitochondrial function.

(a) Oxidative growth phenotype of RRP43 and tetRRP43 strains grown in the presence and absence of doxycycline (0.125 μg ml⁻¹). Equal amounts of serial dilutions (10⁵, 10⁴, 10³, 10²) of cells from exponentially grown cultures were spotted onto YP plates supplemented with 2% glucose (left panel) or with 3% glycerol (right panel). The growth was scored after 3 days of incubation at 28 °C. (b) Cytochrome spectra of RRP43 and tetRRP43 strains grown in the presence and absence of doxycycline (0.125 μg ml⁻¹). The peaks at 550, 560 and 602 nm (vertical bars) correspond to cytochromes c, b and aa₃, respectively. The height of each peak relative to the baseline of each spectrum is an index of cytochrome content. (c) Respiratory activity of yeast RRP43 and tetRRP43 strains grown in the absence and in the presence of doxycycline (0.125 μg ml⁻¹). Wild-type RRP43 and tetRRP43 mutant strain were grown in YP medium supplemented with 0.6% glucose. Respiratory rates were normalized to the wild-type strain grown in the presence of doxycycline. (d) Mitochondrial protein synthesis was performed on RRP43 and tetRRP43 strains in the presence and absence of doxycycline (0,125 μg ml⁻¹), after 15 min of incubation with ³⁵S. Equivalent amounts of protein were separated by SDS–PAGE on a 17.5% polyacrylamide gel, transferred to a nitrocellulose membrane and analyzed by autoradiography.

Figure 5. RT-PCR studies in patient myoblasts, control myoblasts and human oligodendroglia cells and immunohistochemistry and immunoblotting for myelin basic protein (MBP) in human oligodendroglia cells.

(a) Quantitative PCR of MBP, PMP22 and EIF2B2 in patient and control myoblasts. The gene expression of MBP shows significant increase (fold change>2.99, P=0.0055) while no other gene expression is altered. C. M., control myoblasts; P. M., P1-V:10 myoblasts. (b) Relative quantification of MBP expression in human myoblasts treated with EXOSC8 siRNA. Downregulation of EXOSC8 resulted in increased expression of MBP and MOBP, but not in the non-AU-rich control genes PMP22 and EIF2B2. C. M.,control myoblasts; CNTR., control; PP. M., P1-V:10 myoblasts;. (c) RT-PCR detected >100-fold increase of MBP mRNA in human oligodendroglia cells after EXOSC8 downregulation, but no significant change was observed in mRNAs of the non AU-rich control genes PMP22 and EIF2B2. Unpaired t-test was used for statistical analysis. Statistically significant changes are marked with *. Error bars show standard deviation of three experimental repetitions. (d) Quantitative PCR of SMN1, BICD2 and IGHMBP2 in patient and control fibroblasts. The gene expression of SMN1 shows significant increase (~fourfold change, P=0.01984) while no other gene expression is altered. C. F., control fibroblasts; P. F., P3-II:1 fibroblasts. (e) Quantification of ARE-containing and non-ARE-containing genes in human fibroblasts after 3 days EXOSC8 downregulation detected significant increase in SMN1 and MBP. (f) Immunoblotting confirmed that EXOSC8 siRNA downregulation resulted in severe depletion of this protein in both myoblasts (M) and (g) fibroblasts (F). (h) Immunoblotting (repeated three times) confirmed that siRNA downregulation of EXOSC8 resulted in increased MBP protein levels in human oligodendroglia cells. CNTR., control; NT siRNA, non-targeting siRNA. (i) Quantification of MBP protein expression in differentiated oligodendroglia cells after EXOSC8 downregulation. (j) Double immunolabelling of differentiated human oligodendroglia cells for MBP (green channel) and EXOSC8 (red channel). Left column: non-targeting control transfected cells. NT siRNA, non-targeting siRNA. EXOSC8 localized to the nucleus and MBP is distributed throughout the cytosol in EXOSC8 ablated cells. Right column: EXOSC8 staining confirms successful siRNA-mediated downregulation and increased MBP expression. Scale bar, 20 μm.

Figure 6. Knock down of the zebrafish exosc8.

(a) Morphology of live embryos at 48 hpf injected with 12 ng splice-blocking exosc8 antisense MO at the 1–2 cell stage. From left to right: un-injected control embryo, exosc8 MO injected mild, moderate and severe phenotype, respectively. Mild phenotype: slightly curved tail; moderate phenotype: C-shaped; severe category: abnormally formed, very short tail, cardiac oedema, and small, misshapen or missing eyes. (b) RT–PCR analysis: analysis of exosc8 transcripts from embryos injected with the splice-blocking exosc8 MO which targets the splice donor site of exon 2. Using primers in exons 1 and 4, RT–PCR yielded several additional bands in MO injected embryos originating from mis-spliced transcripts. Wild-type transcript is still present in embryos injected with 4.5 ng of MO, but only a trace of wild-type product is left in embryos injected with 12 ng of MO, therefore 12 ng of MO was used in subsequent experiments. wt, wildtype. (c) Relative distribution of the exosc8 morphant phenotypes described above. External morphology described in a and cranial nerve abnormalities displayed in panel d are both taken into account in this categorization; exosc8 MO was injected in three independent experiments into embryos of the Tg(islet-1:GFP) strain and a total of 262 MO injected embryos were evaluated by light and fluorescent microscopy. (d) Brain abnormalities of exosc8 morphants (Tg(islet-1:GFP) strain) from the different phenotypes at 48 hpf: dorsal views of GFP positive neurons in the midbrain and the hindbrain. These transgenic zebrafish embryos express GFP in cranial motor and sensory neurons and in the efferent neurons for the lateral line and the vestibule-acoustic nerves. Normal cranial neuron structure and development was detected in embryos of the normal and mild category. Disrupted neuronal structure in the severe and in a proportion of the moderate categories; GFP positive cells are scattered, no clear structures are visible and overall GFP expression is reduced.

Figure 7. MBP and acetylated tubulin staining after knock down of the zebrafish ortholog: exosc8.

(a) exosc8 MO injected larvae were analyzed for expression of AU-rich mRNAs during 48 hpf by real-time PCR. At 16 hpf expression of mbp1 and plp1 was increased in embryos with an abnormal phenotype, with a similar increase in mbp2 observed at 24 hpf. Despite this initial increase, by 48 hpf there is a dramatic decrease in mbp1 and plp1 expression in larvae with a moderate and severe phenotype. Each bar or severity group at different timepoints represents a number of 15–20 embryos. Statistically significant changes (P<0.05) are marked with *. Unpaired t-test was used for statistical analysis. Error bars represent s.d. of three experimental repeats. (b) Un-injected control larvae and exosc8 MO injected larvae were analyzed for myelination at 96 hpf. Larvae were stained with antibodies against the zebrafish MBP and against acetylated tubulin. Left column: overlay, MBP staining in red, acetylated tubulin staining in green; middle column: MBP staining; right column: acetylated tubulin. Top row: tail of control larva: motor axons in each somite are clearly visible and myelinated at 96 hpf. Second row: tail of MO injected larva with a moderate phenotype: the spinal cord is curved and has an irregular structure. Motor axons in the somites are either very short (arrowhead) and thin or missing completely and are not MBP-positive. Third row: un-injected control larva, posterior lateral line, intact myelin. Bottom row: exosc8 MO injected larva with moderate phenotype: the lateral line is present (green acetylated tubulin signal) but the myelination of its neurons is interrupted (arrowhead). Scale bar, 100 μm. (c) Myelin staining of the lateral line was studied in control un-injected and exosc8 MO injected zebrafish larvae at 4.5 dpf. Representative images of the analyzed embryos are shown on the top (scale bar, 0.25 mm) and transverse sections of the embryos are shown on the bottom. In the control larvae the myelinated lateral line is present at both sides (white arrowheads). However, no myelination of the lateral line was detected in the exosc8 MO injected larvae (white arrowheads). Higher magnifications are shown in the upper right hand corners. Scale bar, 100 μm. (d) Representative EM pictures of the myelin sheath at the lateral line in un-injected and exosc8 MO injected zebrafish larvae at 4 dpf. Black arrows indicate the myelin sheet around the axon. Scale bar, 500 μm.

Figure 8. Simultaneous knock down of exosc8/mbpa/p53 in zebrafish.

(a) Representative graph shows the survival rate of un-injected, mpba+p53, exosc8+p53 and combined mbpa+exosc8+p53 MO injections of the Tg(islet-1:GFP) strain at 48 hpf (summary of 3 experiments). The survival rate increased after triple MO injections compared with exosc8+p53 knockdown embryos. (b) Top row: morphology of the severe embryos at 48 hpf injected with exosc8+p53 and exosc8+mbpa+p53. Bottom row: brain abnormalities of exosc8+p53 and exosc8+mbpa+p53 morphants (Tg(islet-1:GFP) strain) at 48 hpf. Dorsal views of GFP positive neurons in the midbrain and the hindbrain indicate abnormal brain structures in severe exosc8+p53 morphants. Improved cranial neuron structure was detected in severe mbpa+exosc8+p53 MO injected embryos. The same result has been reproduced in three separate experiments with equal MO doses, on the same clutch of embryos.