アイテム詳細

要旨

Adaptation of organisms to continuously changing environments includes the generation of genic novelty in their genomes through mechanisms such as de novo gene evolution, duplication, fusion, lateral gene transfer, etc. De novo gene evolution is a mechanism, wherein new gene functions can evolve from previously non-coding sequences, which are essentially random stretches of nucleotides. Several studies have explored the role of such random sequences as templates for evolutionary innovation. This included a systematic study, where a library of random coding sequences was expressed in Escherichia coli and differential growth was measured to assess fitness effects of individual sequences. Each random sequence from the library was categorized into negative, positive or neutral based on its change in abundance in the population across time. In this thesis, I analyze the effects of individual clones derived from this screen. In order to study effects of random sequences on the fitness of the host, I cloned representative variants from each category into E. coli strains using a multicopy plasmid vector. In the first part of the thesis, I demonstrate that expression of negative random peptides confers a fitness disadvantage (deleterious) in E. coli, followed by a growth recovery. Upon further investigation, I find that these peptides can elicit a stress response in the host instantaneously upon expression. The highly deleterious phenotype can thus be compensated in the host. In addition, I was able to isolate suppressor-of-phenotype clones. Re-sequencing of the suppressors together with each of the ancestor clones helped identify interaction partners for the deleterious peptides. In the second part, I show two mechanisms that the host uses to adapt to deleterious peptide expression: (a) plasmid copy number control by inactivation of the pcnB gene and (b) expression control through inactivation of the LacI inducer binding domain. In the third part of the thesis, I show that the positive random peptides confer competitive fitness advantage only under stressful conditions, for example, an elevated temperature. In conclusion, I show that random sequences indeed affect fitness of the host possibly through targeting specific genes or proteins. This study provides experimental evidence on how random sequences could serve as drivers of de novo gene evolution.