Endonuclease V (EndoV) is highly conserved from prokaryotes to eukaryotes. Endonuclease V in E. coli functions in repair of DNA damages caused by deamination of adenine and guanine in particular. E. coli EndoV also recognizes substrates with discontinuous or distorted DNA helices like base mismatches, loop/hairpins, flaps and junctions, suggesting that EndoV has specificity for DNA with an irregular DNA structure. Thermotoga maritima EndoV (TmaEndoV) shows similar endonucleolytic activity as that of E. coli EndoV. The human EndoV (hEndoV) has lost its ability to recognize and cleave deaminated substrates, and may have developed into a structure-specific DNA binding protein. The structure of EndoV with branched and distorted DNA has not been solved so the molecular details of the interaction of EndoV with different DNA structures are still unknown.
The human Slx1-Slx4 complex has been shown to act as a endonuclease with specific activity on holiday junction DNA. Slx1-Slx4 is highly conserved in eukaryotes. A different family of structure-specific endonucleases is the human Gen1 and yeast Yen1, which also resolves Holliday Junctions. Gen1 and Yen1 are homologues with similar domain organization. No structure of Slx1-Slx4, Gen1 or Yen1 has yet been solved to reveal the molecular details of their interaction with DNA.
Together, the human EndoV, Slx1-Slx4 and Gen1 represent three different protein families of structure-specific endonucleases. In this thesis we present work done with EndoV, Slx1-Slx4 and Gen1/Yen1.
In this thesis, crystallization of TmaEndoV with 3-way junction, base mismatch and hairpin DNA has been done. Two new crystal structures of TmaEndoV have been solved; one of them contains single stranded DNA and the other is the free protein. The new EndoV-DNA complex structure was obtained with the 3-way junction DNA, however, only single-stranded DNA is observed in the electron density map calculated from the diffraction data. The DNA substrate was made using 12-mer oligonucleotides, and the binding of EndoV to this substrate thus seems to affect the hybridization of the DNA in a way that results in single-strand DNA binding only.
Further, different expression plasmids of human Slx1 and yeast Slx1 have been tested. Human Slx1 (hSlx1) expression plasmid shows good expression of the protein, but purification shows that the protein is unstable in free form. An expression plasmid of the nuclease domain of hSlx1 was designed, but the MBP-Slx1 fusion protein is not cleavable. Co-purification of Slx1 and Slx4 did not improve the yield of free Slx1. Slx1 from the yeasts S. pombe and S. cerevisiae shows no expression when fused to MBP.
Expression tests of human Gen1 and S. cerevisiae Yen1 shows no expression of an MBP fusion protein.