De novo Transcriptome Assembly and Analysis of Placental Tissue from Undernourished, Control Fed and Obese Pregnant Ewes

The incidence of maternal malnourishment and obesity is significant and increasing in populations throughout the world, and has been linked to altered fetal growth rate, and increased obesity, insulin resistance and cardiovascular disease in offspring. To investigate the placental mechanisms involved, an ovine model of maternal malnutrition was utilized, where ewes received either 50% (nutrient restricted, NR) or 150% (obese, OB) of the control (C) diet (100% of nutrient requirements) through midgestation (d75). In this model, fetal growth and development is markedly decreased (~30%) or increased (~30%) at necropsy on d75 in NR and OB ewes, respectively compared to fetuses of C ewes. Placental tissue was collected from individual placentomes (functional unit of fetal:maternal exchange), snap frozen and stored at -80˚C.  To assess the transcriptome dynamics involved in obesity and malnourishment, RNAseq was used to compare gene expression differences in NR, OB and C ewes. In absence of a complete or well annotated Ovis aries reference genome, we sequenced and de novo assembled a sheep transcriptome (104,293 transcripts) from 8.14Gbps of single (66 million) and paired (42 million) Illumina reads, computationally predicted protein coding transcripts (22,666) with Bovine models and annotated functional domains of these with Hidden Markov Model strategies and Gene Ontology mapping. We realigned the RNAseq reads for each sample to the custom reference transcriptome to estimate readcount-based expression levels and identified genes differentially expressed (DE) in each of the 3 pairwise comparisons (NR v C—155 DE genes;  OB v C—66 DE genes; OB v NR—216 DE genes). Our work represents a detailed collection and annotation of high quality transcripts of the Ovine model and presents functional enrichment and gene-specific analysis of obesity and malnourishment.