The primary goal of this project was to complete a comparative survey of the genome sequences of the set of morphologically cryptic species of the Paramecium aurelia assemblage, a group of ciliated protozoans with ancient roots. As the first study of this sort in a natural assemblage of unicellular eukaryotes, this project has greatly expanded our understanding of the mechanisms of genome evolution, providing a complement to the much richer availability of observations on multicellular species. Notably, the two-genome duplication scenario for the P. aurelia complex is identical in form to (and potentially more ancient than) that postulated for the origin of morphological diversification of the vertebrates. Yet, the ciliate assemblage exhibits near absolute stasis at the morphological level, despite underlying changes at the molecular level. Thus, the results of this project are of considerable relevance to ongoing speculation regarding the connection between genome duplication and morphological innovation.

Intellectual merit. By establishing the complete history of all ancestral gene copies over a finely dissected phylogeny, it has been possible to evaluate the patterns of preservation vs. demise of various functional classes of duplicate genes. The analyses also reveal the temporal patterns of gene loss that eventually lead to the acquisition of new equilibrium genomic states in the descendant taxa. This shows, in turn, how reciprocal losses of sister genes in related taxa has an effect similar to gene relocation, thereby resulting in the passive evolution of reproductive isolation among sister taxa. Our results reveal a remarkable simplicity of Paramecium genome architecture, most notably introns never exceeding 30 base pairs in length, and intergenic regions typically shorter than 100 base pairs. Not only do these very large cells have some of the most streamlined genomes in all of eukaryotes, but their chromosomal gene order has remained nearly constant for over 500 million years. 

            Using information on gene expression, tests of several key hypotheses on the evolution of duplicate genes have revealed that ancestral gene-expression level is the primary determinant of the fate of a duplicate gene. This has led to the development of a model for evolution of expression level after a gene duplication, related to earlier ideas on qualitative subfunctionalization. Under this model, duplicated genes initially share identical biochemical functions and are retained because of dosage constraints. However, mutations in regulatory regions and trans-acting factors gradually allow divergence of expression level between the two copies. The expression level of each individual duplicated gene can evolve neutrally so long as the joint expression of the pair remains roughly constant. However, once a high level of imbalance is reached, the copy with the lowest expression level contributes a small enough fraction to total expression that selection no longer opposes its loss. This model provides an explanation as to why the most highly expressed duplicate genes (which are furthest from the absorbing barrier of low expression) exhibit the highest longevities.

            Owing to the extreme simplicity of the intergenic regions associated with Paramecium genes, it has been possible to identify candidate transcription-factor binding sites (TFBSs, both repressors and activators) by bioinformatics methods. These results show that the distances of such elements from their client genes are highly correlated with gene expression, enabling us to indirectly validate repressor and enhancer effects even prior to functional analysis.

Broader impact. A central goal o...

Please report errors in award information by writing to: awardsearch@nsf.gov.