Molecular evolution and ecology of fleas and flea-associated microbial organisms in light of horizontal gene transfer
MetadataShow full item record
As an infamous group of blood-sucking ectoparasitic insects, fleas (order Siphonaptera) pose a significant threat to the health of humans, and other animals. Not only are their bites irritating, but fleas are also a major link in continued outbreaks of plague, typhus, cat-scratch disease and other important epidemics. This is because they are vectors of several groups of pathogenic bacteria, such as Yersinia, Rickettsia and Bartonella. These highly transmittable microbes interact with their arthropod vectors and mammalian reservoirs on ecological and physiological levels and have long evolutionary associations with their vectors, which influences their genetic makeup. Specifically, when two or more groups of bacteria coexist, there is a notable chance that they exchange genetic material, a process termed horizontal gene transfer (HGT). This process plays a major role in the evolution of microbial genomes by introducing new functions and altering existing functions, including for instance the ability of infecting host cells and resistance to antibiotics. Together, the above mentioned bacteria, insects and mammals form an interesting three-level, network-like host-parasite system with highly sophisticated dynamics. In order to efficiently control vector-borne, and in this case flea-transmitted diseases, it is important to understand the ecology and evolution of Siphonaptera themselves, as well as their microbial associates. The former can be assessed in the framework of a biodiversity survey and molecular phylogenetics, while the latter calls for new methods and tools considering horizontal bacterial gene flow. In particular, the identification of HGT remains a developing field. While conventional phylogenetic approaches can be used to resolve the evolutionary history of individual gene families, rapid, robust and reliable methods for genome-wide scans of HGT candidates are still lacking. My dissertation is composed of three independent projects serving the same general goal. In the first project, I studied the evolution of fleas, and conducted a comprehensive phylogenetic study based on a global sampling of more than 250 flea DNA samples, representing the vast majority of the known familial biodiversity of this insect order. A phylogenetic tree with deep-level species resolution and statistical support was reconstructed, providing answers to long-standing debates regarding the evolution of basal flea families and the relationships among major flea lineages. Using cutting edge statistical approaches and fossil-based calibrations, the ages of major splits in the evolutionary history of fleas were estimated for the first time, pointing toward an origination in the Early Cretaceous and the emergence of major clades in the Late Cretaceous, before the K-Pg extinction event. The host association and geographical distribution records of current flea species were collected and compiled, based on which the ancestral states at these major splits were reconstructed. My results suggest that the common ancestor of modern fleas was already associated with Theria. Both, Metatheria (marsupials) and Eutheria (placental mammals) are likely ancestral hosts, in spite of the current dominance of rodents as major hosts of extant fleas. Moreover, the current geographic distribution and timing of radiations make it likely that fleas started on the Gondwana supercontinent in the Southern Hemisphere. Taken together, my study provides robust and valuable insights into the evolution and ecology of early Siphonaptera. In the second project, I studied the pattern of HGTs in the genomes of Bartonella, a major group of microbial pathogens vectored by fleas. A genome-wide preliminary search based on BLAST was conducted to narrow down the range of putative HGT-derived genes, followed by explicit phylogenetic analyses on these candidate gene families. The result reveals a series of interesting cases of gene losses and horizontal transfers in the phospholipid biosynthesis pathway. Specifically, the gpsA gene, which encodes the NAD(P)H-dependent glycerol-3-phosphate dehydrogenase, was lost in the ancestral Bartonella lineage, but was horizontally reacquired independently in three extant lineages. The sources of these heterogeneous copies were phylogenetically located with the clades of the mammalian pathogen Helicobacter and the arthropod symbionts Arsenophonus and Serratia, respectively. This rare phenomenon suggests an ancestral, possibly obligate intracellular lifestyle of Bartonella, followed by subsequent functional adaptations toward more flexible routes of transmission. My study was accompanied by an experimental biodiversity survey of Bartonella symbionts of fleas and bat flies, which illuminated the post-transfer evolution of Bartonella gpsA genes. This study provides insights into the evolution of flea-associated pathogens from a genomic perspective, and presents an interesting case of HGT that plays an important role in shaping the physiological and ecological characteristics of a bacterial group. In the third project, I developed a novel BLAST-based approach for genome-wide surveys of putative HGT events, coded in Perl. Unlike conventional BLAST-based methods, which rely merely on the best match, this method features the statistical analysis of BLAST hit distribution patterns of genomes with phylogenetic consideration of the organisms of interest. I tested this method on both simulated and real-world genomic data. For the latter, I analyzed the genomic data of Rickettsia, another important group of flea-borne pathogens. The comparison of my results to previous analyses suggest that this method effectively isolates genes with a putative horizontal origin from the majority of genes which are vertically inherited. Compared to conventional approaches, HGTector is notably insensitive to stochastic effects that can lead to false positives and false negatives. Moreover, it is rapid, exhaustive, consumes very limited computational resource, and is a useful addition to the toolbox of evolutionary biologists interested in horizontal aspects of microbial evolution. An automated pipeline was created to implement this approach and was made publicly available at: https://github.com/DittmarLab/HGTector .