Adaptations in nucleus accumbens neuron subtypes mediate negative affective behaviors in fentanyl abstinence: Gene expression analysis
For RNA sequencing, only samples with RNA integrity numbers >8 were used. 6 samples/sex/drug/cell-type were submitted for RNA sequencing at the UMSOM Institute for Genome Sciences (IGS) and processed as described previously (Engeln et al., 2020). Libraries were prepared from 10 ng of RNA from each sample using the Smart-Seq v4 kit (Takara). Samples were sequenced on an Illumina HiSeq 4000 with a 75 bp paired-end read. An average of 64–100 million reads were obtained for each sample. Reads were aligned to the mouse genome (Mus_musculus.GRCm38) using TopHat (version 2.0.8; maximum number of mismatches = 2; segment length = 30; maximum multi-hits per read = 25; maximum intron length = 50,000). The number of reads that aligned to the predicted coding regions was determined using HTSeq. We applied two strategies to characterize gene expression changes in these data. First, we sought to identify individual genes with significant gene expression changes following abstinence from fentanyl. We used limma-trend to fit log2-normalized read counts per million to a linear model and tested for significant effects of fentanyl in each cell type, separately in males and in females, as well as treating sex as a covariate. Second, we characterized the effects of fentanyl on gene co-expression networks. As a starting point for this analysis, we used limma-trend to select the set of genes that exhibited a nominally-significant effect (p-value < 0.05) of fentanyl, sex, or cell type. We then used weighted gene co-expression network analysis (WGCNA) to characterize co-expressed modules among these genes, separately within D1 and D2 MSNs. Module detection was performed with the blockwiseModules() function with power = 10, corType = ‘bicor’, networkType = ‘signed’, minModuleSize = 25, reassignThreshold = 0, mergeCutHeight = 0.25, minMEtoStay = 0, and otherwise default parameters. We tested for differential expression of module eigengenes with limma, using posthoc contrasts to estimate effects of fentanyl while adjusting for sex differences, as above. Hub genes for each module were defined by calculating the Pearson correlation between the eigengene and each gene in the module.