Differential gene expression in response to drought treatment in California blue oak (Quercus douglasii) seedlings

Background/Question/Methods

Drought may pose a significant abiotic stress in plant populations experiencing increased temperatures as climates change. The ability of tree populations to respond to drought may depend on their level of standing genetic variation in molecular pathways that regulate water stress.  To elucidate the genes involved in these pathways, transcriptional profiles can be compared between water-deprived and control plants to reveal differentially expressed genes.  Here, we used RNAseq to generate transcriptional profiles for California blue oak (Quercus douglasii) seedlings growing under two water regimes in the greenhouse.  Our goals were 1) to identify genes involved in response to drought and 2) to test whether there is variation in gene expression among maternal lines.  To do this, we subjected 24 seedlings from 6 maternal sources to either control or dry-down treatments, and sequenced RNA from leaf tissue collected before and after each treatment using 4 lanes of Illumina single-end 50 base pair sequencing.  We mapped reads to a recently published Q. lobata reference transcriptome. We then used principal components analysis to detect overall differences in gene expression patterns and DESeq2 to identify genes that were differentially expressed before and after drought stress. 

Results/Conclusions

Among our 48 samples (24 seedlings x 2 time points), we obtained over 820 million reads, of which 80% aligned to over 79,000 contigs in the reference transcriptome. Principal components analysis of read counts per contig revealed more clustering by maternal genotype than by either time or treatment, indicating that gene expression depends more on genetic background than on environmental influences.  However, DESeq2 analysis identified a panel of 2,783 genes that were significantly differentially expressed before and after drought treatment.  We further identified 15 genes whose differential gene expression in response to drought depended on maternal line, indicating that variable strategies for response to water limitation may exist within a population.  We are currently analyzing expression differences between time points in control seedlings in order to identify genes involved in growth.  These growth genes will be subtracted from the panel of differentially expressed genes in drought-stressed seedlings, as a means of isolating genes involved primarily in the water stress pathway. This work identifies genes involved in regulating water stress in the long-lived tree species, Quercus douglasii, and provides a list of candidates that can be used in future studies examining adaptive genetic variation across the species range.