Subunit Interactions in Regulation and Catalysis of Site-Specific Recombination

Wenwieser, Sandra Verena Corinna Tina (2001) Subunit Interactions in Regulation and Catalysis of Site-Specific Recombination. PhD thesis, University of Glasgow.

Abstract

The role of the 2,3'-interface as an interdimer interface required for synapsis was investigated by creating a double mutant R2A/E56K and combining this with the activating mutations D102Y/E124Q. The effects of multiple 2,3' mutations on regulation of recombination by sites II/III were found to be additive. Mutation of two residues at the 2,3'-interface (R2A/E56K) abolished all detectable activity (except binding), even when the mutations are targeted exclusively to site I. Nevertheless, when targeted to sites II and III, R2A/E56K could fulfil some of the regulatory functions of WT resolvase. Mutation of the same residues in an activated resolvase (R2A/E56K/D102Y/E124Q) did not block catalysis at site I, and had only a partial effect on regulation by sites II and III. Thus, the most critical function of the 2,3'- interface may be to support activation of site I-bound resolvase by sites II and III. While it is clear that the 2,3' interaction contributes to the stability of the synapse, these data imply that it may not be the main architectural interface. This role of the 2,3'-interface supports a new model of the synapse (Sarkis et ai, in preparation). A potential candidate interface for synapsis of resolvase dimers was the region around residue D/E 102. Mutations of residues in this region have resulted in activated resolvases, which show reduced dependence on the accessory sites. Residues 96-105 of Tn3 resolvase were replaced with the homologous sequence of Hin, thereby introducing six amino acid substitutions. This resulted in an activated resolvase which showed cleavage and some recombination activity in the absence of sites II and III. This was the first accessory site independent resolvase that did not also carry activating mutations around E124, demonstrating that mutation of the 102 region is sufficient. Furthermore, the rate of res x res recombination was increased with this mutant. When the corresponding region of gammadelta resolvase was replaced with Hin sequence, the mutant protein was highly active, recombining crossover sites at a similar rate in the presence or absence of accessory sites. Nevertheless, it was clearly established that both Tn3R (96-105 Hin) and gammadeltaR (96-105 Hin) readily synapsed sites II/III, trapping three negative supercoils. Thus, these mutants recombined crossover sites in the absence of sites II/III, yet were synapsis-proficient when sites II/III were present. Both mutants also showed significant supercoil-independent cleavage and recombination activity, but were nevertheless stimulated by negative supercoiling, and by sites II/III. A further mutant, gammadeltaR (96-105 Hin) E124Q, generated large amounts of covalent cleavage complexes as predicted. In contrast to the earlier mutants, it was extremely supercoil and accessory site independent, making it an ideal candidate for crystallisation trials for the cleavage intermediate. Targeting experiments to determine whether mutation in the 102 region disrupted synapsis with WT resolvase were difficult to design and interpret because of the highly activated properties of these (96-105 Hin)-swapped resolvases. gammadeltaR (96-105 Hin) was also used to investigate whether this region of Hin was involved in direct contacts with the Fis/enhancer complex. When assayed with a substrate analogous to the natural Hin substrate (with res crossover sites replacing the hix sites), a small bias towards inversion over resolution products (approximately 2:1) was observed in the presence of Fis and HU. This bias may be indicative of the formation of an invertasome structure. However, the bias observed was not limited to resolvase carrying the (96-105 Hin) substitution and was also seen with gammadeltaR E102Y/E124Q. The analogous experiment to look for possible contact surfaces was performed with a resolvase/Hin hybrid containing a Hin DNA binding domain designed to bind res binding sites. A variety of WT and activated resolvase versions of this hybrid were constructed and were found to be catalytically active. Additionally, gammadeltaR E102Y/E124Q Hin/Res DBD mutant was active with a site I x site I substrate. However, preliminary results have shown no effect of Fis and HU on the reaction rate or products indicating that the Hin DNA binding domain is not involved in direct contacts with the Fis/enhancer complex. Models of strand exchange differ radically in their predictions of which resolvase interfaces must be disrupted in order to bring the half-sites together in the recombinant configuration. In order to investigate this, cysteine residues were introduced into resolvase so that three different components of the dimer interface could be disulphide or chemically crosslinked. The mutants initially constructed (D95C/A113C, T73C/A115C (96-105 Hin) and M106C (96- 105 Hin)) were fully active when reduced but completely inactive when oxidised (except for binding). These crosslinked derivatives were therefore defective in a step prior to cleavage, thus making it impossible to investigate strand exchange requirements. Nevertheless, resolvases crosslinked in the trans and E helix interfaces were found to be proficient in all of the preceding functions including binding to res, accessory site synapsis, and activation of resolvase at site I. This rules out a domain-swapping mechanism for synapsis of sites II/III. (Abstract shortened by ProQuest.).

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Additional Information: Adviser: Marshall Stark
• Keywords: Genetics
• Date of Award: 2001
• Depositing User: Enlighten Team
• Unique ID: glathesis:2001-74962
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 27 Sep 2019 14:58
• Last Modified: 27 Sep 2019 14:58
• URI: https://theses.gla.ac.uk/id/eprint/74962