The invention of oxygenic photosynthesis has forever changed the face of the Earth by producing the oxygen present in our atmosphere. It is thought that this unique metabolism arose from the bacterial phylum, Cyanobacteria. Moreover, this group has significantly contributed to eukaryotic diversity via endosymbiosis, as a cyanobacterium is considered to be the progenitor of the original plastid organelle, more commonly recognized as chloroplasts in plants. The antiquity of these events, dearth of convincing methodologies, and lack of conclusive evidence all contribute to the considerable challenges we face in understanding the timing of these major evolutionary and geological transitions. In order to address these problems, I have employed various techniques focusing on improving our understanding of the role of cyanobacteria and photosynthesis in shaping the world we have today.
Evolutionary relationships are difficult to reconstruct due to phylogenetic noise (e.g. horizontal gene transfer and homoplasy), resulting in uncertainty in our ability to build accurate phylogenetic trees. In order to address this issue, fifty-four strains of cyanobacteria were chosen for genome sequencing based on improving the phylogenetic coverage of the phylum. Not only does the diversity-driven and phylum-level approach identify many novel genes, but it also clarifies the phylogenetic placement of various cyanobacterial subclades, protein families, and endosymbiosis events.
The timing of ancient events, such as primary endosymbiosis events, has primarily been dependent on the fossil record. This is problematic as microfossils are difficult to interpret and assign to extant lineages. Conversely, molecular clock methods have been just as widely varying. We devised a new approach to increase the amount of dating information incorporated into molecular clock analyses, improving the accuracy of the predicted dates. We date the plastid and mitochondrial endosymbiosis events to approximately 900 and 1200 million years ago, respectively.
Finally, I focus on the protein evolution of a specific protein crucial to photosynthesis: RuBisCO. Here, I use ancestral sequence reconstruction methods to predict, synthesize, and characterize ancestral versions of RuBisCO. We show that ancestral RuBisCOs have lower rates of carboxylation, reflective of the high CO2 and low O2 Precambrian atmosphere.