In search of non-photosynthetic Cyanobacteria

Abstract

One of the major evolutionary events that occurred on our planet is the establishment of organisms capable of performing oxygenic photosynthesis. This changed the earth’s atmosphere from an anoxic to an oxic environment which likely contributed to the development of more complex organisms. Cyanobacteria are the only known prokaryotes capable of performing oxygenic photosynthesis and until recently, it was believed that all Cyanobacteria carry out this process. With the advent of culture-independent molecular techniques, a number of basal lineages of Cyanobacteria have been detected and classified as 4C0d-2 and ML635J-21. Many representatives of these lineages have been found in aphotic environments, raising the possibility that they are non-photosynthetic. The main aim of this thesis was to use metagenomics to obtain genomes belonging to the cyanobacterial basal lineage 4C0d-2, determine if they contain the photosynthetic apparatus required for oxygenic photosynthesis and whether they should be classified as Cyanobacteria or as a separate phylum. In Chapter Two, amplicon pyrosequencing was used to identify environments that contain members of the basal lineage, 4C0d-2. Positive habitats were shotgun sequenced and six genomes were extracted from 4C0d-2 populations using differential coverage binning; three from koala faeces, one from human faeces, one from a lab-scale (EBPR), and one from a full-scale (UASB). No photosynthetic genes were identified in any of the 4C0d-2 genomes suggesting that this lineage is indeed non-photosynthetic. Genome-based phylogenetic trees confirmed that 4C0d-2 shares a common ancestor with photosynthetic cyanobacteria. An independent study by Di Rienzi et al (2013) concluded that 4C0d-2 is a sister phylum of the Cyanobacteria for which they proposed the name Melainabacteria. Based on the robust phylogenetic association and a number of inferred common traits between the two groups, I proposed that the Melainabacteria should be reclassified as a class within the Cyanobacteria, comprising four orders represented by genome sequences; Gastranaerophilales, Obscuribacterales, Caenarcaniphilales and Vampirovibrionales. During analysis of Melainabacteria 16S rRNA genes, the sequence of a possible cultured representative, Vampirovibrio chlorellavorus, was discovered (after which I named one of the orders above). In the 1970’s, V. chlorellavorus was observed preying upon the microalga, Chlorella vulgaris and was initially classified as a Bdellovibrio (member of the Deltaproteobacteria) based on its predatory nature and cell shape. In Chapter Three, the DNA from 36 year-old lyophilised cells of V. chlorellavorus and C. vulgaris were extracted and the genome of V. chlorellavorus was shotgun sequenced and assembled into a near-complete draft genome. Genome-based trees confirmed that V. chlorellavorus is a member of the Melainabacteria and not the Deltaproteobacteria, expanding the number of known phyla containing predatory bacteria to five. The molecular machinery used by V. chlorellavorus for predation was inferred from the annotated genome. Representatives from the order Obscuribacterales were detected in a sample collected from intact permafrost (palsa) as part of another study. In Chapter Four, two genomes from this order were extracted from ultra-deep metagenomic sequencing and metabolic reconstructions were created to provide an in-depth insight into their functionality. Metatranscriptomic analysis of one palsa sample was performed to determine which genes are actively expressed by the permafrost Obscuribacterales. The findings presented in this thesis provide a useful basis for understanding the newly discovered non-photosynthetic cyanobacterial lineage, the Melainabacteria, and increases the number of known phyla containing predatory bacteria. For example, the Melainabacteria genomes may be used to understand the evolutionary origins of oxygenic photosynthesis, design probes for visualisation or identify potential media for culturing.