Large-scale estimates of cellular origins of mRNAs: enhancing the yield of transcriptome analyses - PubMed (original) (raw)

Large-scale estimates of cellular origins of mRNAs: enhancing the yield of transcriptome analyses

Etienne Sibille et al. J Neurosci Methods. 2008.

Abstract

Gene expression profiling holds great promise for identifying molecular pathologies of central nervous system disorders. However, the analysis of brain tissue poses unique analytical challenges, as typical microarray signals represent averaged transcript levels across neuronal and glial cell populations. Here we have generated ratios of gene transcript levels between gray and adjacent white matter samples to estimate the relative cellular origins of expression. We show that incorporating these ratios into transcriptome analysis (i) provides new analytical perspectives, (ii) increases the potential for biological insight obtained from postmortem transcriptome studies, (iii) expands knowledge about glial and neuronal cellular programs and (iv) facilitates the generation of cell-type specific hypotheses. This approach represents a robust and cost-effective "add-on" to transcriptome analyses of the mammalian brain. As this approach can be applied post hoc, we provide tables of ratios for analysis of existing mouse and human brain datasets.

PubMed Disclaimer

Figures

Figure 1

Figure 1. Brain area and cross-species correlations of WM/GM ratios

A. Correlation graphs of WM/GM ratios for human/human (∼17,000 genes), human/mouse (∼6,100 orthologous genes), and mouse/mouse (∼21,000 genes) brain area comparisons. Axis scales, −6< Log2(WM/GM) <6. See also Fig.3A. B. Pearson correlation factors for area and species comparisons of WM/GM ratios. Significance of correlations, p<e−7 for all r.

Figure 2

Figure 2. The WM/glial content of GM samples is mostly subject-dependent

Dots represent the average percentile of deviation in WM content per sample across the two PFC areas in cohort 1. Most samples displayed variable levels of WM content within GM samples, although values agreed well across two closely related PFC areas, suggesting a biological determination of glial content within subjects. Three GM samples display larger discrepancies in their WM content across areas (Black dots). (r=0.51, p=0.002 all samples; r=0.83, p< e−7, grey samples). Hashed bars represent line of similar values (Slope =1).

Figure 3

Figure 3. Using WM/GM ratios highlights the extent and cell-type specificity of age-related transcriptome changes

A. Age-effected genes superimposed on WM/GM correlation graph in human PFC. Age-affected genes (red) represented only 14% of the most neuronal- (Black) and 6% of the glial- (Blue) genes and were evenly distributed across the spectrum of WM/GM ratios. B. Functional analysis based on WM/GM transcript enrichment separates the contribution of glia and neurons to molecular aging. Left: “All genes combined” Gene Ontology (GO) analysis of altered gene expression in aging PFC. The top 40 most affected gene groups are presented out of ∼900 groups investigated. Right: Analysis of the same dataset using WM- or GM-enriched gene pools identified two distinct sets of functions. Altered glial-related gene groups (Yellow; upper right) mostly concerned cellular defenses, whereas most age-affected neuronal gene groups (Gray; lower right) related to synaptic functions and signal transduction. The top 25 most affected gene groups are presented for each analysis. White bars indicated functions affected in both cellular populations. Details in Erraji-BenChekroun et al., 2005.

References

    1. Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, et al. Gene ontology: tool for the unification of biology. The Gene Ontology Consortium. Nat Genet. 2000;25:25–9. - PMC - PubMed
    1. Cotter D, Mackay D, Landau S, Kerwin R, Everall I. Reduced glial cell density and neuronal size in the anterior cingulate cortex in major depressive disorder. Arch Gen Psychiatry. 2001;58:545–53. - PubMed
    1. Dennis G, Jr., Sherman BT, Hosack DA, Yang J, Gao W, Lane HC, et al. DAVID: Database for Annotation, Visualization, and Integrated Discovery. Genome Biol. 2003;4:3. - PubMed
    1. Erraji-BenChekroun L, Underwood MD, Arango V, Galfalvy HC, Pavlidis P, Smyrniotopoulos P, et al. Molecular aging in human prefrontal cortex is selective and continuous throughout adult life. Biological Psychiatry. 2005;57:549–58. - PubMed
    1. Galfalvy HC, Erraji-Benchekroun L, Smyrniotopoulos P, Pavlidis P, Ellis SP, Mann JJ, et al. Sex genes for genomic analysis in human brain: internal controls for comparison of probe level data extraction. BMC Bioinformatics. 2003;4:37. - PMC - PubMed

Publication types

MeSH terms

Substances

Grants and funding

LinkOut - more resources