'Pseudo' domains in phage-encoded DNA methyltransferases

Nature. 1991 Aug 15;352(6336):645-8. doi: 10.1038/352645a0.

Abstract

5-Cytosine-DNA-methyltransferases, which are found in many organisms ranging from bacteriophages to mammals, transfer a methyl group from S-adenosylmethionine to the carbon-5 of a cytosine residue in specific DNA target sequences. Some phage-encoded methyltransferases methylate more than one sequence: these enzymes contain several independent target-recognizing domains each responsible for recognizing a different site. The amino-acid sequences of these multispecific methyltransferases reveal that some enzymes in addition carry domains that do not contribute to the enzymes' methylation potential, but strongly resemble previously identified target-recognizing domains. Here we show that introducing defined amino-acid alterations into these inactive domains endows these enzymes with additional methylation specificities. Gel retardation analysis demonstrates that these novel methylation specificities correlate with the acquisition of additional DNA-binding potential of the proteins.

MeSH terms

  • Amino Acid Sequence
  • Bacteriophages / enzymology
  • Binding Sites
  • DNA (Cytosine-5-)-Methyltransferases / chemistry*
  • DNA (Cytosine-5-)-Methyltransferases / metabolism
  • DNA-Binding Proteins / chemistry
  • DNA-Binding Proteins / metabolism
  • DNA-Cytosine Methylases / chemistry
  • DNA-Cytosine Methylases / metabolism
  • Molecular Sequence Data
  • Structure-Activity Relationship
  • Substrate Specificity

Substances

  • DNA-Binding Proteins
  • DNA-Cytosine Methylases
  • DNA (Cytosine-5-)-Methyltransferases