To gain insight into how mutant huntingtin (mHtt) CAG repeat length modifies Huntington's disease (HD) pathogenesis, we profiled mRNA in over 600 brain and peripheral tissue samples from HD knock-in mice with increasing CAG repeat lengths. We found repeat length-dependent transcriptional signatures to be prominent in the striatum, less so in cortex, and minimal in the liver. Coexpression network analyses revealed 13 striatal and 5 cortical modules that correlated highly with CAG length and age, and that were preserved in HD models and sometimes in patients. Top striatal modules implicated mHtt CAG length and age in graded impairment in the expression of identity genes for striatal medium spiny neurons and in dysregulation of cyclic AMP signaling, cell death and protocadherin genes. We used proteomics to confirm 790 genes and 5 striatal modules with CAG length–dependent dysregulation at the protein level, and validated 22 striatal module genes as modifiers of mHtt toxicities in vivo.
Access optionsAccess options
Subscribe to Journal
Get full journal access for 1 year
only $18.75 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
Gene Expression Omnibus
Gene Expression Omnibus
We thank PsychoGenics for help in breeding the knock-in allelic series and dissecting the tissues as part of a contract research agreement with CHDI. The research was supported by CHDI Foundation, Inc. HD research in the Yang laboratory is also supported by NINDS US National Institutes of Health grants (R01NS074312, R01NS049501 and R01NS084298). X.W.Y is also supported by the David Weill fund from Semel Institute, the Carol Moss Spivak Scholarship in Neuroscience from the Brain Research Institute at UCLA, and the Leslie Gehry Brenner Prize from the Hereditary Disease Foundation. We acknowledge the support of the NINDS Informatics Center for Neurogenetics and Neurogenomics (P30 NS062691).
Integrated supplementary information
Differential expression statistics in 2-, 6-, and 10-month striatum, cortex and liver. This table contains 9 sheets, each sheet corresponding to one tissue/time point combination. In each sheet, rows correspond to genes. The first 3 columns identify the gene and the other columns provide differential expression statistics for two-group comparisons (Q80, Q92, Q111, Q140, Q175 vs. Q20) and for the association test with Q as a numeric variable.
Differential expression between Q175 mice and controls in tissue survey. Each of the sheets in this table provides differential expression statistics for one of the 14 tissues in the tissue survey. In each sheet, the first 3 columns identify the gene and the other columns provide differential expression statistics. The last sheet, named ‘Correlation of differential expression Z statistics’, provides the correlations among vectors of differential expression Z statistics that are displayed in graphical form in Figure 1D.
Summary of network analysis results. Sheets ‘Striatum, (Cortex, Liver) module membership’ give, for each gene, its assigned module label and color, meta-analysis Z statistics for module membership in all modules, and module membership (also known as kME) in all modules at each time point. These tables can be used as a resource in two ways: Given a module, one can identify consensus hub genes (genes with highest module membership Z statistics) in the module as candidates for further follow-up; and conversely, given an interesting gene, one can check whether it is a hubgene in any of the modules. The sheets labeled ‘Striatum (Cortex, Liver) module-Q association’ provides a summary of association between module eigengenes (Methods) and genotype. Analogously to differential expression testing, we test association of module eigengenes with Q viewed as a continuous variable as well as in two-group comparisons of Q80, Q92, Q111, Q140 and Q175 vs. Q20. Test statistics reported in these tables include correlation, Student t-test statistics, Kruskal-Wallis test statistics, as well as descriptive statistics (means, standard errors, numbers of observations etc.). We also report the meta-analysis Z and significance statistics that pool test results across the three ages.
Annotation of top 18 striatal modules from network analyses.
Gene ontology analysis of top modules. Brain modules with Meta Z score greater than 5 were assessed for enrichment using DAVID Gene Ontology Functional Annotation Clustering (Huang et al., 2009). Tabs correspond to individual modules, with color representing sign of the Meta Z score (green, negative; red, positive).
IPA canonical pathway analysis of top modules. Brain modules with Meta Z score greater than 5 were assessed for pathway enrichment using Ingenuity Pathway Analysis Canonical Pathways (Qiagen, Redwood City, CA; http://www.qiagen.com/ingenuity). Tabs correspond to individual modules, with color representing sign of the Meta Z score (green, negative; red, positive).
Top module enrichment of top modules in HDinHD BrainLists anRicher function. Brain modules with Meta Z score greater than 5 were assessed for enrichment using the HDinHD anRicher function limited to BrainLists (http://www.hdinhd.org). This probes datasets related to brain region and cell types, disease, and aging using the userListEnrichment function (Miller et al., 2011). Tabs correspond to individual modules, with color representing sign of the Meta Z score (green, negative; red, positive).
Preservation of association between module genes and genotype or HD status in independent data. This table provides, in a text form, data that are shown in Figure 3. Specifically, for each of the 18 selected striatum and cortex modules, the Table shows weighted mean correlation with genotype (mouse data) or HD status (human data) of module genes across 24 test data sets, as well as the corresponding p-values.
Overview of HD-related literature gene expression data sets used for validation.
Genes that change consistently in allelic series and human data. In each sheet (Striatum, Cortex), each row corresponds to a gene that is consistently and significantly expressed in 6-month allelic series and human data. For striatum we report the 6-month allelic series striatum and the human CN data sets by Durrenberger et al. and Hodges et al. Each striatal gene satisfies the following criteria: FDR<0.05 in the allelic series striatum, FDR<0.1 in each of the human data sets, and same sign of fold change across all 3 data sets. For the cortex, we report the allelic series 6-month cortex, BA4 and BA9 data by Hodges et al., and PFC and VC data from the Harvard Brain Tissue Resource Center (Zhang et al., 2014). Each cortical gene satisfies the following criteria: FDR<0.05 in the allelic series cortex, FDR<0.1 in at least 3 of the 4 of the human data sets, and same sign of fold change in the allelic series cortex and at least 3 of the 4 human data sets.
Enrichment of selected striatum and cortex modules in informative marker sets. This table contains gene marker sets that show nominally significant (p<0.05) enrichment in selected striatum and cortex modules. For the striatum, the marker sets include top 100 ABA striatal and cortex markers, several D1 and D2-specific gene sets, cadherins/protocadherins, and genes determined to change significantly in HD patients using laser capture microdissection (LCM). For cortex modules, we tested for enrichment in top 100 ABA cortex and striatum markers.
Preservation of association between cell death genes in striatum M7 and genotype or HD status in literature data. This table provides, in a text form, data shown in Figures 5C-E. Specifically, this table shows weighted mean correlation of cell death genes in Striatum M7 with genotype (mouse data) or HD status (human data) of module genes across 24 test data sets, as well as the corresponding p-values.
Proteomic label-free quantification (LFQ) data and sample information.
Numbers of significantly differentially abundant proteins across all genotypes. For each comparison, the table lists the number of significantly (FDR<0.1) differentially abundant proteins, as well as the number of significantly differentially abundant proteins whose mRNA is also significantly (FDR<0.1) associated with the genotype variable, as well as the corresponding hypergeometric overlap p-values.
Summary statistics of protein network modules. This table includes association of module eigen-proteins with genotype and a summary of functional enrichment analysis.
Enrichment of CAG-dependent mRNA modules in differentially abundant proteins.
Genes tested in Drosophila HD model.
Summary of the validation in Drosophila HD model. For each tested gene, columns give gene identification, module number from our WGCNA analysis in the striatum, Allele type (LOF, loss of function, shRNA, shRNA knock-down; O, overexpression) and modifier effect (E, enhancer; S, suppressor).
Drosophila HD model p and F values for statistics.
Sample numbers across tissues, genotypes and time points. The individual sheets in this table provide sample numbers at each genotype and time point for the fully profiled tissues, the tissue survey at 6 months, and the proteomic data.
About this article
Neuroscience Bulletin (2018)