- 10.00: Trial lecture
- 12.00: Public defence
The ordinary opponents are:
- First opponent: Associate Professor Åsmund Røhr Kjendseth, Norwegian University of Life Sciences (NMBU)
- Second opponent: Associate Professor Elena Papaleo, Technical University of Denmark (DTU)
- Leader of the committee: Professor Lisbeth Charlotte Olsen, OsloMet
The leader of the public defense is Associate Professor Hege Tunsjø, OsloMet.
The main supervisor is Associate Professor Ole Herman Ambur, OsloMet. The co-supervisors are Scientist, PhD Kristian Alfsnes, Norwegian Institute of Public Health (NIPH) and Research Scientist, PhD Stephan Alfons Frye, Oslo University Hospital.
Thesis abstract
Bacteria heavily rely on a mechanism dubbed horizontal gene transfer (HGT), which can be seen as swapping of “cheat codes” for survival – recipes for survival, proven through countless of generations. Naked DNA, floating in the environment, is taken up by bacterial cells and incorporated into their genetic makeup (genome) if it stems from the same species (transformation).
Such DNA fragments often contain genes that serve as replacements for genes that are damaged in the recipient cell and might also contain antibiotic resistance genes or other genes that compromise human health. However, bacteria do not blindly grab random DNA from the environment; they are selective.
In two bacterial families that contain many species that pose threats to human health, Neisseriaceae and Pasteurellaceae, HGT is regulated by DNA sequence motifs of length 9-12 DNA base pairs, known as DNA Uptake Enhancing Sequences (DUES).
These are found in thousands of copies within species of these families and must be present in naked environmental DNA for DNA uptake to occur. If naked DNA in the environment contains one such DUES, receptors on the bacterial cell surface will recognize the DUES and bind to it, leading to DNA uptake and transformation (natural competence).
In two of the studies implemented in this PhD thesis, the artificial intelligence (AI) structural modeling tool AlphaFold3 (AF3) was a common denominator. The first study sought to better understand the interaction between the cell surface receptor in Neisseriaceae (ComP) and DUES using AF3 and two other AI tools to predict their 3D structures. AF3 was also used in the second study to infer the long-sought cell surface receptor in Pasteurellaceae (ComN).
With high probability, my AF3 models proved that ComN preferentially binds DUES over other DNA sequences. As such, my work may represent a game-changer, validating AI’s ability to solve complex biological challenges that previously required costly and time-consuming laboratory work.
The third study in this PhD research focused on whether certain gene categories are particularly enriched in DUES and also different functions of DUES in Neisseriaceae and Pasteurellaceae genomes.
This study used hundreds of bacterial genomes to show that DUES are highly constrained by evolution, often found in regions of the genome related to its maintenance and positioned specifically to minimize their metabolic cost.
The findings in this PhD research might aid in the combat against human pathogens in Neisseriaceae and Pasteurellaceae. Therapeutic agents could be developed to block ComP/ComN from binding DUES, thus stopping the spread of antibiotic resistance genes.
