Table 2.
Study | Samples | Platform | Measure of differential expression (no. differentially expressed) | Main findings |
---|---|---|---|---|
SMA | ||||
Garcia et al. [230] | Smn-null and wild type drosophila lavae (n = 2 each group). RNA extracted when Smn-null lavae first display motor phenotype | Illumina HiSeq 2000 | Gene level: Compared with wild type; Fisher's exact test. Exon level: Compared with wild type, DEXSeq, P < 0.05 (2153 splicing changes); MISO, Bayes factor > 100 (2484 splicing changes) | Comparison of gene level analysis with modENODE data [231] consistent with developmental arrest in the Smn-null animals; splicing changes exceeded normal developmental changes |
Zhang et al. [232] | Microdissected motor neurons and glial cells from spinal cord of SMN-deficient and wild-type mice at postnatal day 1 before MN pathology develops (n = 2 each group) | Illumina HiSeq 2000 | Compared with wild type. Gene level:, DESeq, P < 0.05, FC > 2 (Motor neurons: 138↑ and 110↓; Glial cells: 125↑ and 87↓). Exon level: MISO, Bayes factor > 10 (Motor neurons: 104; glial cells: 86 splicing changes) | Dysregulation of genes implicated in synaptogenesis. Minimal overlap of transcriptome changes between cell types despite broadly similar genes expressed |
See et al. [125] | Zebrafish embryos at 48h post-fertilization injected morphilino vs. SMN, Gemin2 or scrambled (n = 6–7 each group) | Custom zebrafish microarray [233] | Compared with scrambled, t-test, P < 0.05, fold change > 2 (1545 transcripts) | Down-regulation of Neurexin2a which is involved in synaptic function. Knockdown of Neurexin2a phenocopied the SMA model in zebrafish |
Acsadi et al. [234] | NSC34 cells transfected with shRNA vs. SMN or scrambled | StellARray™ microarray NE0100-MM96 | NSC34 cells, vs. scrambled fold change > 2 (5 genes). Note that custom array contains only genes linked to neurodegeneration. | SNCA down-regulation highest fold change. Change confirmed in peripheral and CNS tissue from SMA patients. SNCA is implicated in synaptic function and PD. |
C9orf72-linked ALS | ||||
Donnelly et al. [153] | RNA extracted from motor cortex, fibroblasts and iPSNs from C9orf72+ ALS patients and controls (n = 2–4 each group). Followed by transfection of C9orf72+ iPSNs with ASO vs. C9orf72 | Human Affymetrix 1.0 ST Exon microarray. | Compared with control, t-test P < 0.05. (Fibroblasts: 487 genes↑ and 1238 genes↓; iPSNs: 984 genes↑ and 2314 genes↓; motor cortex: 1870 genes↑ and 5114 genes↓) | Significant number of differentially expressed genes in common between C9orf72+ iPSNs and C9orf72+ motor cortex. ASO treatment normalized gene expression in C9orf72+ iPSNs. |
Sareen et al. [154] | RNA extracted from fibroblasts and iPSNs from C9orf72+ ALS patients and controls (n = 4 each group). Followed by transfection of C9orf72+ iPSNs with ASO vs. C9orf72 or scrambled ASO | Illumina HiSeq. | Compared with control, t-test P < 0.05 | Differentially expressed genes in iPSNs functionally enriched for synaptic transmission. ASO vs. C9orf72 partially corrected gene expression in C9orf72+ iPSNs. |
Lagier-Tourenne et al. [155] | RNA extracted from fibroblasts of C9orf72+ ALS patients, non-C9orf72 ALS patients and controls (n = 4 each group). Followed by transfection of C9orf72+ samples with ASO vs. C9orf72 | Illumina Hi-Seq | C9orf72+ compared with controls, FDR < 0.05 (122 genes↑, 34 genes↓) | Treatment with ASO vs. C9orf72 failed to reverse expression signature of C9orf72+ fibroblasts. |
Ismail et al. [235] | RNA extracted from lymphoblastoid cell lines derived from C9orf72+ ALS patients, non-C9orf72 ALS patients and controls (n = 10–16 each group) | Human Affymetrix Genome U133 Plus 2.0 microarray | Compared with controls. PPLR [236], PLikeValue < 0.05 (C9orf72+ vs. control: 319 genes; C9orf72+ vs. non-C9orf72: 294 genes) | Differentially expressed genes functionally enriched for NF-κB activity. Top differentially expressed gene CXCL10 down-regulation in C9orf72+ samples; this gene is controlled by NF-κB and is a positive prognostic marker in sporadic ALS [237]. |
TDP-43-liked ALS | ||||
Huang et al. [188] | Astrocytes purified from (GFAP)-tTa/TRE-TDP-43M337v double transgenic rats. RNA extracted after 3, 4 or 6 days of induction of mtTDP-43 expression | Microarray | Compared with baseline FC > 2 at ≥1 time point (449 genes) and progressive change in FC over induction time points | Induction of TDP-43M337V expression altered expression of secreted factors; in particular, expression of neurotrophic factors increased and expression of neurotoxic factors decreased. |
Arnold et al. [186] | Transgenic mice produced which express TDP-43Q331K, or TDP-43wild-type throughout CNS. RNA was extracted from cortices and spinal cords of 2-month-old mice and compared with nontransgenic mice. | Affymetrix ‘A = chip’ microarray. | Compared with nontransgenic. Absolute separation score < 0.3, q < 0.05 (Cortex: TDP-43wild-type 824 exons alternatively spliced; TDP-43Q331K 1195 exons; Spinal cord: TDP-43wild-type 4462 exons, TDP-43Q331K 4399 exons) | Differentially spliced exons in mutant, but not wild-type transgenic mice enriched for known TDP-43 binding sites. Differentially spliced exons in TDP-43Q331K which overlapped between cortex and spinal cord enriched for synaptic function |
Highley et al. [189] | RNA extracted from fibroblasts derived from mtTDP-43 ALS patients, mtSOD1 ALS patients, sporadic ALS patients and controls (n = 3–6 each group) | Human Affymetrix Exon 1.0 ST microarray | Compared with controls, ancova P < 0.01 (mtTDP-43: 1070 genes↑ and 433 genes↓, 12 280 exons; mtSOD1: 6335 exons; sporadic ALS: 1140 exons) | Functional enrichment in mtTDP-43 differentially expressed/spliced genes for categories related to RNA processing. Alternative splicing much more abundant in mt-TDP-43 vs. other ALS patients |
ALS, amyotrophic lateral sclerosis; ASO, antisense oligonucleotide; FC, fold change; FDR, false discovery rate; iPSNs, induced pluripotent stem cell-derived motor neurons; SMA, spinal muscular atrophy.