Degree Type


Date of Award


Degree Name

Doctor of Philosophy


Genetics, Development and Cell Biology

First Advisor

Dan Voytas


Transposable elements have great potential to restructure the genome of their hosts by mutating genes through insertion, imposing regulation on adjacent genes, and causing recombination-based genome rearrangements. The movement of transposable elements, therefore, could have two consequences to the host genome: under normal conditions, high mobility of transposable elements will be harmful to the integrity of the host genome, whereas under stress conditions, transposition could benefit host cells by creating genetic variations that offer selective advantages. McClintock hypothesized that the host cell and transposable elements have co-evolved diverse mechanisms to mutually benefit their survival.The yeast retrotransposon, Ty5, has proven to be a great model to understand the integration site preference of retrotransposons. Ty5 preferentially integrates into heterochromatin at the telomeres and silent mating loci. It has been shown that Ty5 integration preference is directed by the interaction between a small domain of Ty5 integrase (the targeting domain, TD) and Sir4, a heterochromatic protein. In this thesis, we demonstrated that serine 1095 in TD is phosphorylated by the host cell and that this phosphorylation is essential for interacting with Sir4 and subsequently mediating Ty5 integration into heterochromatin. In the absence of phosphorylation, created for example by substituting serine 1095 with alanine, Ty5 becomes a potent mutagen. More than half of the insertions generated by phosphorylation-defective TD mutants occur in coding sequences. Therefore, the yeast cell actively controls Ty5's mutagenic potential through posttranslational modification, and this minimizes deleterious consequences of transposition. Furthermore, TD phosphorylation is reduced under stress conditions, specifically starvation for amino acids, nitrogen, or fermentable carbon, allowing the Ty5 integrations to occur in euchromatin. This result is consistent with McClintock's hypothesis that mobile elements reshape host genomes as an adaptive response to environmental stress.The crosstalk between Ty5 and the host cell is further evidenced by the regulation of heterochromatic silencing by Ty5. It has been shown that the expression of Ty5 TD disrupts silencing at telomeres and HM loci, and this anti-silencing activity depends on TD's ability to interact with Sir4. Single amino acid substitutions that abolish TD interaction with Sir4 no longer break transcriptional silencing. We showed that TD breaks silencing by mediating Sir4 turnover, which is dependent on 1) Ubc4, an ubiquitin conjugating enzyme, 2) Ris1, a RING type ubiquitin ligase and 3) the proteasome pathway. We, for the first time, demonstrate that the RING finger domain of Ris1 has direct ubiquitination activity, and this activity requires the intact form of the RING finger. Interestingly, we found that a domain located in the C-terminus of Esc1, which is functionally equivalent to TD, also possesses Ubc4- and Ris1-dependent anti-silencing activity. Like TD, this Esc1 domain interacts with Sir4 and directs Ty5 integration when it is substituted for TD within integrase. Our results suggest that Ty5 has the potential to regulate the dynamics of heterochromain in a fashion similar to Esc1, a host cellular protein that regulates heterochromatin under certain events such as cell cycle progression and stress. We further assessed the regulation of Ty5 transposition by the cell cycle and DNA damage to understand the interrelationship between Ty5, the host cell and environmental conditions. We found that Ty5 integration is not inhibited in yeast cells arrested in G1 phase by the mating pheromone alpha factor, although transposition of Ty1 and Ty3 is repressed by alpha factor treatment. In addition, we demonstrated that Ty5 cDNA integration is restricted upon induction of DNA damage by HU and MMS, but that Ty5 cDNA recombination is not affected. By performing genetic analysis, we confirmed that depletion of Sml1, which results upon DNA damage, is the cause for the Ty5 integration deficiency. The mechanism by which Sml1 regulates Ty5 integration is under investigation. When yeast cells lack Esc1 or Ris1, or experience DNA damage, Ty5 still appears to predominantly integrate into heterochromatin. These conclusions are based on the characterization of a handful of Ty5 integration sites. To explore the possibility that Ty5 integration patterns are modestly changed under these conditions, we developed an important new approach for studying global Ty5 integration patterns that uses linker mediated PCR followed by 454 sequencing and bioinformatic analysis. We sequenced more than 300,000 PCR amplicons and mapped 20,000 unique insertions and 85,000 multi-hit insertions onto the yeast genome. These insertions were derived from both haploid and diploid strains. Our mapping data demonstrate that Ty5 integrates throughout domains of heterochromatin at the telomeres, silent mating loci and rDNA. This integration pattern, however, is not random: most insertions fall within intergenic regions. Furthermore, the rare Ty5 insertions that occur in euchromatin also occur in intergenic regions in both haploid and diploid strains, demonstrating that the bias is not due to selection. This distribution pattern supports the safe haven hypothesis, namely that Ty5 is very-well adapted to life in the yeast genome because under normal growth conditions, it preferentially inserts into gene-poor region to minimize the potential deleterious effects on host genome integrity.


Copyright Owner

Jiquan Gao



Date Available


File Format


File Size

113 pages