Differential allelic representation (DAR) identifies candidate eQTLs and improves transcriptome analysis.

In comparisons between mutant and wild-type genotypes, transcriptome analysis can reveal the direct impacts of a mutation, together with the homeostatic responses of the biological system. Recent studies have highlighted that, when the effects of homozygosity for recessive mutations are studied in non-isogenic backgrounds, genes located proximal to the mutation on the same chromosome often appear over-represented among those genes identified as differentially expressed (DE). One hypothesis suggests that DE genes chromosomally linked to a mutation may not reflect functional responses to the mutation but, instead, result from an unequal distribution of expression quantitative trait loci (eQTLs) between sample groups of mutant or wild-type genotypes. This is problematic because eQTL expression differences are difficult to distinguish from genes that are DE due to functional responses to a mutation. Here we show that chromosomally co-located differentially expressed genes (CC-DEGs) are also observed in analyses of dominant mutations in heterozygotes. We define a method and a metric to quantify, in RNA-sequencing data, localised differential allelic representation (DAR) between those sample groups subjected to differential expression analysis. We show how the DAR metric can predict regions prone to eQTL-driven differential expression, and how it can improve functional enrichment analyses through gene exclusion or weighting-based approaches. Advantageously, this improved ability to identify probable eQTLs also reveals examples of CC-DEGs that are likely to be functionally related to a mutant phenotype. This supports a long-standing prediction that selection for advantageous linkage disequilibrium influences chromosome evolution. By comparing the genomes of zebrafish (Danio rerio) and medaka (Oryzias latipes), a teleost with a conserved ancestral karyotype, we find possible examples of chromosomal aggregation of CC-DEGs during evolution of the zebrafish lineage. Our method for DAR analysis requires only RNA-sequencing data, facilitating its application across new and existing datasets.

The Use of Zebrafish in Transcriptome Analysis of the Early Effects of Mutations Causing Early Onset Familial Alzheimer's Disease and Other Inherited Neurodegenerative Conditions.

The degree to which non-human animals can be used to model Alzheimer's disease is a contentious issue, particularly as there is still widespread disagreement regarding the pathogenesis of this neurodegenerative dementia. The currently popular transgenic models are based on artificial expression of genes mutated in early onset forms of familial Alzheimer's disease (EOfAD). Uncertainty regarding the veracity of these models led us to focus on heterozygous, single mutations of endogenous genes (knock-in models) as these most closely resemble the genetic state of humans with EOfAD, and so incorporate the fewest assumptions regarding pathological mechanism. We have generated a number of lines of zebrafish bearing EOfAD-like and non-EOfAD-like mutations in genes equivalent to human PSEN1, PSEN2, and SORL1. To analyze the young adult brain transcriptomes of these mutants, we exploited the ability of zebrafish to produce very large families of simultaneous siblings composed of a variety of genotypes and raised in a uniform environment. This "intra-family" analysis strategy greatly reduced genetic and environmental "noise" thereby allowing detection of subtle changes in gene sets after bulk RNA sequencing of entire brains. Changes to oxidative phosphorylation were predicted for all EOfAD-like mutations in the three genes studied. Here we describe some of the analytical lessons learned in our program combining zebrafish genome editing with transcriptomics to understand the molecular pathologies of neurodegenerative disease.

Differential allelic representation (DAR) identifies candidate eQTLs and improves transcriptome analysis.

In comparisons between mutant and wild-type genotypes, transcriptome analysis can reveal the direct impacts of a mutation, together with the homeostatic responses of the biological system. Recent studies have highlighted that, when homozygous mutations are studied in non-isogenic backgrounds, genes from the same chromosome as a mutation often appear over-represented among differentially expressed (DE) genes. One hypothesis suggests that DE genes chromosomally linked to a mutation may not reflect true biological responses to the mutation but, instead, result from differences in representation of expression quantitative trait loci (eQTLs) between sample groups selected on the basis of mutant or wild-type genotype. This is problematic when inclusion of spurious DE genes in a functional enrichment study results in incorrect inferences of mutation effect. Here we show that chromosomally co-located differentially expressed genes (CC-DEGs) can also be observed in analyses of dominant mutations in heterozygotes. We define a method and a metric to quantify, in RNA-sequencing data, localised differential allelic representation (DAR) between groups of samples subject to differential expression analysis. We show how the DAR metric can predict regions prone to eQTL-driven differential expression, and how it can improve functional enrichment analyses through gene exclusion or weighting of gene-level rankings. Advantageously, this improved ability to identify probable eQTLs also reveals examples of CC-DEGs that are likely to be functionally related to a mutant phenotype. This supports a long-standing prediction that selection for advantageous linkage disequilibrium influences chromosome evolution. By comparing the genomes of zebrafish (Danio rerio) and medaka (Oryzias latipes), a teleost with a conserved ancestral karyotype, we find possible examples of chromosomal aggregation of CC-DEGs during evolution of the zebrafish lineage. The DAR metric provides a solid foundation for addressing the eQTL issue in new and existing datasets because it relies solely on RNA-sequencing data.

The evolved divergence of γ-secretase-susceptibility of homologous proteins Ngfrb and Nradd in zebrafish.

OBJECTIVE: NGFR/p75NTR and NRADD/NRH proteins are closely related structurally and are encoded by genes that arose from a duplication event early in vertebrate evolution. The transmembrane domain (TMD) of NGFR is cleaved by γ-secretase but there is conflicting data around the susceptibility to γ-secretase cleavage of NRADD proteins. If NGFR and NRADD show differential susceptibility to γ-secretase, then they can be used to dissect the structural constraints determining substrate susceptibility. We sought to test this differential susceptibility. RESULTS: We developed labelled, lumenally-truncated forms of zebrafish Ngfrb and Nradd and a chimeric protein in which the TMD of Nradd was replaced with the TMD of Ngfrb. We expressed these in zebrafish embryos to test their susceptibility to γ-secretase cleavage by monitoring their stability using western immunoblotting. Inhibition of γ-secretase activity using DAPT increased the stability of only the Ngfrb construct. Our results support that only NGFR is cleaved by γ-secretase. Either NGFR evolved γ-secretase-susceptibility since its creation by gene duplication, or NRADD evolved to be refractory to γ-secretase. Protein structure outside of the TMD of NGFR is likely required for susceptibility to γ-secretase.

Zebrafish Chromosome 14 Gene Differential Expression in the fmr1 h u2787 Model of Fragile X Syndrome.

Zebrafish represent a valuable model for investigating the molecular and cellular basis of Fragile X syndrome (FXS). Reduced expression of the zebrafish FMR1 orthologous gene, fmr1, causes developmental and behavioural phenotypes related to FXS. Zebrafish homozygous for the hu2787 non-sense mutation allele of fmr1 are widely used to model FXS, although FXS-relevant phenotypes seen from morpholino antisense oligonucleotide (morpholino) suppression of fmr1 transcript translation were not observed when hu2787 was first described. The subsequent discovery of transcriptional adaptation (a form of genetic compensation), whereby mutations causing non-sense-mediated decay of transcripts can drive compensatory upregulation of homologous transcripts independent of protein feedback loops, suggested an explanation for the differences reported. We examined the whole-embryo transcriptome effects of homozygosity for fmr1 h u2787 at 2 days post fertilisation. We observed statistically significant changes in expression of a number of gene transcripts, but none from genes showing sequence homology to fmr1. Enrichment testing of differentially expressed genes implied effects on lysosome function and glycosphingolipid biosynthesis. The majority of the differentially expressed genes are located, like fmr1, on Chromosome 14. Quantitative PCR tests did not support that this was artefactual due to changes in relative chromosome abundance. Enrichment testing of the "leading edge" differentially expressed genes from Chromosome 14 revealed that their co-location on this chromosome may be associated with roles in brain development and function. The differential expression of functionally related genes due to mutation of fmr1, and located on the same chromosome as fmr1, is consistent with R.A. Fisher's assertion that the selective advantage of co-segregation of particular combinations of alleles of genes will favour, during evolution, chromosomal rearrangements that place them in linkage disequilibrium on the same chromosome. However, we cannot exclude that the apparent differential expression of genes on Chromosome 14 genes was, (if only in part), caused by differences between the expression of alleles of genes unrelated to the effects of the fmr1 h u2787 mutation and made manifest due to the limited, but non-zero, allelic diversity between the genotypes compared.