Interaction of the E-coli DNA G : T-mismatch endonuclease (vsr protein) with oligonucleotides containing its target sequence

Citation
Dp. Turner et Ba. Connolly, Interaction of the E-coli DNA G : T-mismatch endonuclease (vsr protein) with oligonucleotides containing its target sequence, J MOL BIOL, 304(5), 2000, pp. 765-778
Citations number
52
Categorie Soggetti
Molecular Biology & Genetics
Journal title
JOURNAL OF MOLECULAR BIOLOGY
ISSN journal
00222836 → ACNP
Volume
304
Issue
5
Year of publication
2000
Pages
765 - 778
Database
ISI
SICI code
0022-2836(200012)304:5<765:IOTEDG>2.0.ZU;2-X
Abstract
The Escherichia coli vsr endonuclease recognises G:T base-pair mismatches i n double-stranded DNA and initiates a repair pathway by hydrolysing the pho sphate group 5' to the incorrectly paired T. The enzyme shows a preference for G:T mismatches within a particular sequence context, derived from the r ecognition site of the E. coil dcm DNA-methyltransferase (CC[A/T]GG). Thus, the preferred substrate for the vsr protein is (CT[A/T]GG), where the unde rlined T is opposed by a dG base. This paper provides quantitative data for the interaction of the vsr protein with a number of oligonucleotides conta ining G:T mismatches. Evaluation of specificity constant (k(st)/K-D; k(st) = rate constant for single turnover, K-D = equilibrium dissociation constan t) confirms vsr's preference for a G:T mismatch within a hemi-methylated dc m sequence, i.e. the best substrate is a duplex (both strands written in th e 5'-3' orientation) composed of CT[A/T]CG and C-5Me[T/A]GG. Conversion of the mispaired T (underlined) to dU or the d(5Me)C to dC gave poorer substra tes. No interaction was observed with oligonucleotides that lacked a G:T mi smatch or did not possess a dcm sequence. An analysis of the fraction of ac tive protein, by "reverse-titration" (i.e. adding increasing amounts of DNA to a fixed amount of protein followed by gel-mobility shift analysis) show ed that less than 1% of the vsr endonuclease was able to bind to the substr ate. This was confirmed using "competitive titrations" (where competitor ol igonucleotides are used to displace a P-32-labelled nucleic acid from the v sr protein) and burst kinetic analysis. This result is discussed in the lig ht of previous in vitro and in vivo data which indicate that the MutL prote in may be needed for full vsr activity. (C) 2000 Academic Press.