Ranaweera, L. T. 2026. Expression evolution in spatial and temporal contexts in plants and modeling the regulation of gene expression. Dissertation, Michigan State University, East Lansing, MI.

Deciphering the relationship between variations in DNA sequences and the resulting diversity of phenotypes often termed the genotype-phenotype map, remains a foundational challenge in evolutionary biology and quantitative genetics. While the connections between changes in protein-coding sequences and altered protein functions are relatively clear, the impact of genetic variations within non-coding regions, such as intergenic or intronic sequences, is less straightforward to establish. To bridge this gap, this dissertation utilizes the study of gene expression dynamics and evolution as a proxy to understand how regulatory divergence and environmental interactions and evolutionary factors such as gene duplications affect observable traits. By integrating computational biology, machine learning, and phylogenetics, this research investigates the regulatory architecture underlying environmental adaptation in switchgrass and the evolutionary dynamics of spatial expression following genome duplication events.

The first portion of this research focuses on the molecular mechanisms underlying climatic adaptation, specifically the transcriptional response to cold stress in switchgrass (Panicum virgatum). Low-land ecotypes possess significantly higher biomass, but lower cold tolerance compared to up-land ecotypes. To understand this disparity, I computationally dissected the temporal cis-regulatory basis of the switchgrass cold response. Using an existing transcriptome dataset, I established machine learning models predictive of cold response based on k-mer sequences enriched in the genic and flanking regions of cold-responsive genes. This approach identified 655 putative cis-regulatory elements (pCREs), with 54 identified as important across all cold treatment time points (30 minutes to 24 hours). Notably, while eight of these pCREs were similar to known cold-responsive elements, the majority were novel sequences. These findings suggest that an amplified, cascading effect in gene expression is regulated by a complex set of previously unknown sequence elements that warrant further functional study.

Building upon the identification of regulatory elements, I further quantified the genetic and regulatory factors shaping cold-responsive transcriptional divergence through the lens of genotype-by-environment (GxE) interactions in the second chapter. By comparing a cold-tolerant upland ecotype (DAC6) and a cold-susceptible lowland ecotype (AP13), I modeled the effects of Genotype (G), Environment (E), and GxE on gene expression. My results indicate that G accounts for the largest proportion of expression divergence (~53%) between ecotypes, while E effects contribute significantly (42%). Using allele-specific expression in F1 hybrids to define cis- and trans-regulatory contributions, I found that cis-regulatory divergence contributes most strongly to parental expression divergence. In contrast, trans-regulatory variation contributes more to expression plasticity, the changes in gene expression across different environments. This study reveals how a relatively small proportion of genes (~5%) exhibiting significant GxE effects likely facilitates adaptation to low-temperature stress, providing potential candite genes for future experimental validations.

Finally, this dissertation to investigate how large-scale genome-wide events, such as genome duplication, contribute to the evolution of gene expression in different tissue types. Using tomato as a model, I integrated pairwise paralog expression comparisons with phylogenetic analyses of ancestral expression states to examine genes retained from the most recent genome triplication event (Sol-WGT). My analysis reveals that ancestral expression states tended to be retained by approximately 66% of Sol-WGT duplicates. However, asymmetrical expression divergence was prevalent among retained duplicate gene pairs, particularly when ancestral genes were at extreme expression levels. We observed, genes that gained expression in fruit tissues were significantly enriched with functions associated with fruit quality. This study demonstrates that the gain or loss of ancestral expression in duplicates is coordinated between tissues derived from similar organs or developmental processes, suggesting a tight spatial regulation and co-evolution of gene expression in related tissue types.

In summary, this dissertation provides a comprehensive exploration of the regulatory architecture that translates genetic variation into phenotypic diversity. By characterizing novel pCREs, quantifying the cis/trans regulatory divergence and GxE interactions, and mapping the spatial evolution of duplicated genes, this work advances our understanding of the evolution of phenotypic diversity of plants. These findings have broader implications for the improvement of high-yield crops under environmental stress and provide critical insights into the long-term evolutionary consequences in plant genomes.

Associated Treatment Areas:

• GLBRC Research Context

Download citation to endnote bibtex