Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2015 Sep;282(18):3556-78.
doi: 10.1111/febs.13358. Epub 2015 Jul 16.

The Sequenced Rat Brain Transcriptome--Its Use in Identifying Networks Predisposing Alcohol Consumption

Free PMC article

The Sequenced Rat Brain Transcriptome--Its Use in Identifying Networks Predisposing Alcohol Consumption

Laura M Saba et al. FEBS J. .
Free PMC article


A quantitative genetic approach, which involves correlation of transcriptional networks with the phenotype in a recombinant inbred (RI) population and in selectively bred lines of rats, and determination of coinciding quantitative trait loci for gene expression and the trait of interest, has been applied in the present study. In this analysis, a novel approach was used that combined DNA-Seq data, data from brain exon array analysis of HXB/BXH RI rat strains and six pairs of rat lines selectively bred for high and low alcohol preference, and RNA-Seq data (including rat brain transcriptome reconstruction) to quantify transcript expression levels, generate co-expression modules and identify biological functions that contribute to the predisposition of consuming varying amounts of alcohol. A gene co-expression module was identified in the RI rat strains that contained both annotated and unannotated transcripts expressed in the brain, and was associated with alcohol consumption in the RI panel. This module was found to be enriched with differentially expressed genes from the selected lines of rats. The candidate genes within the module and differentially expressed genes between high and low drinking selected lines were associated with glia (microglia and astrocytes) and could be categorized as being related to immune function, energy metabolism and calcium homeostasis, as well as glial-neuronal communication. The results of the present study show that there are multiple combinations of genetic factors that can produce the same phenotypic outcome. Although no single gene accounts for predisposition to a particular level of alcohol consumption in every animal model, coordinated differential expression of subsets of genes in the identified pathways produce similar phenotypic outcomes.

Database: The datasets supporting the results of the present study are available at

Keywords: HXB RI rat strains; alcohol consumption; gene expression; neuroinflammation; rat selected lines.


Figure 1
Figure 1. Genes/isoforms differentially expressed between high alcohol consuming and low alcohol consuming selected lines of rats
Genes/isoforms were ranked by p-value from the meta-analysis including all six selected line pairs and the top 10 genes (A) and isoforms (B) are included in the figure. Each row of the heatmap represents a gene/isoform and each column represents a selected line pair. The top line of each box is the log2 difference in expression (high consuming line – low consuming line). The bottom line is the p-value for the difference in expression related to that particular pair. The color of the boxes are based on the log2 difference in expression.
Figure 2
Figure 2. LOD profile of voluntary alcohol consumption in the HXB/BXH recombinant inbred panel
Strain means were used in a marker regression to determine behavioral QTL for voluntary alcohol consumption in the 2-bottle 24-hour access paradigm. Two suggestive (p<0.63) QTL are labels with their location, credible interval, LOD score, and genome-wide p-value. The red line represents the LOD threshold for a suggestive p-value (2.38, genome-wide p-value=0.63). The two insets are more detailed views of the 2 suggestive peaks. Their 90% Bayesian credible intervals are shaded grey. The QTLs are labeled with their location, credible interval, and the number of Ensembl transcripts and Ensembl protein-coding transcripts physically located within each QTL's credible interval.
Figure 3
Figure 3. Connectivity within the co-expression module associated with voluntary alcohol consumption
Each node represents a gene and/or an isoform from the two co-expression modules that were associated with alcohol consumption using a p-value that combined information from the correlation of the eigengene with alcohol consumption and the enrichment of genes/isoforms within module differentially expressed in the rat lines selectively bred for high or low alcohol consumption. The size of the node is weighted based on its intramodule connectivity within the merged co-expression module. Nodes highlighted in yellow represent genes identified in both the gene-level analysis and the isoform-level analysis. The thickness of the line connecting two nodes, i.e., edge, is weighted based on the magnitude of the correlation coefficient between the two genes. Red edges represent a negative correlation and blue edges represent a positive correlation.
Figure 4
Figure 4. Comparison of the transcriptome structure of novel rat transcript across mouse and human
The top box (Rat) is the genomic region, chr12:40,902,059–40,918,309, in the RGSC 5.0/rn5 version of the rat genome. In the rat, the novel transcript is transcribed from the negative strand. The numerical values of the coordinates have been reversed, 40,918,309 bp to 40,902,059 bp, so that the direction of transcription (left to right in the graphic) is consistent across species. In this box, the transcript structure of three transcripts derived from the transcriptome reconstruction using the polyA+-selected RNA is shown as the first series of tracks in black (e.g., GENE_07345.ISO_1). GENE_07346 is the hub gene for the co-expression module (Figure 3). The second series (grey) in this box is the exon organization of GENE_07345 deduced from the PCR product sequence. The third series of tracks within this box indicate the genomic regions in the rat that are orthologous to the A930024E05Rik gene in mouse. The labels on the right are the relevant mouse RefSeq ncRNA ID. The final series of tracks in this box indicates the genomic region in the rat that is orthologous to LOC101928346 in humans. The label on the left is the relevant human RefSeq ncRNA ID. The second box (Mouse) is the genomic region, chr5:122,988,841-123,005,091, in the GRCm38/mm10 version of the mouse genome. The track within this box contains the A930024E05Rik gene as annotated in mouse. Regions that were identified as orthologous to the rat are colored with the same colors used in the Regions Orthologous to Mouse A930024E05Rik in the Rat box above. The third box (Human) is the genomic region, chr12:121579996-121596246, in the GRCh38/hg38 version of the human genome. The track within this box is the LOC101928346 lincRNA annotated in human with the relevant human RefSeq ncRNA IDs on the left. Regions that were identified as orthologous in the rat are colored with the same colors as in the Regions Orthologous to Human LOC101928346 in the Rat box above. It should be noted that the GENE_07346 and orthologous regions in the other two species are located between the Kdm2b and the Orai1 gene sequences. This figure was generated using the UCSC Genome Browser (
Figure 5
Figure 5. Functional relationships among candidate genes for alcohol consumption
These cartoons illustrate the functions of and interactions among the annotated candidate genes for alcohol consumption that are described in more detail in the text, the Supporting Information, and Table 3. Functions and interactions were derived from the Formal Concept Analysis and most candidate genes are expressed in glial cells (astrocytes and/or microglia), and each panel of the figure represents one of the functional categories listed in Table 3. Candidate genes are shown in red.

Similar articles

See all similar articles

Cited by 24 articles

See all "Cited by" articles

Publication types