Brief Description of Oligo-Directed Mutagenesis  

Oligonucleotide-directed mutagenesis, a standard tool of molecular biology, works as follows. Suppose we have a long "template" strand of DNA containing some arbitrary sequence of bases (named A,T,C,G) that includes some site of interest, and we put this DNA into a solution containing many copies of short DNA strands (oligos) that are almost perfect Watson-Crick complements (A binds to T, C binds to G) of the region around the site of interest, except for one or a few bases in the oligo that are mismatched (or possibly inserted or deleted), relative to the site. If conditions are right, the oligonucleotide will still stick to the site, and can serve as a primer, meaning that a polymerase (5) enzyme can attach to and extend the oligo, synthesizing a copy of the long DNA template in the downstream (3') direction from the oligo. (6) Meanwhile, a growing copy of the DNA from some primer lying upstream (5') of the oligo can meet the oligo and then be covalently bonded to it by a ligase (7) enzyme (see this figure). Thus, the mismatched oligo becomes incorporated into a newly-formed long strand of DNA which is almost a perfect negative (8) of the original strand, but differs from it in a specific way that is defined by the base-sequence of the particular mutagenic oligonucleotide that is present. The key feature of the mechanism is that the oligo will bind to, and thus cause miscopying of, only selected sites in the template DNA; at other sites there are just too many mismatches for the pairing to occur at all. For more information about oligonucleotide-directed mutageneis, see [Watson-et-al-92].

  ligase joins       24-base oligonucleotide
  strand growing     having 2 mismatches with    polymerase
  from upstream      the given site on template  extends the
  to the oligo\           |                      oligo in the
               \          |                        / 3' direction
                \  5'------------------------>3'  /
   - - - - - >       GTCAATTGCATGCCACGTATAACT - - - - - - - - >
                     ::::::X::::::::X::::::::
   ...TAGTATGCGGCGTATCAGTTAGCGTACGGTACATATTGACGTGTCGTATGACGTTGA...
3'<--------------------------------------------------------------- 5'
             /             |        |
            /               \       |
  template DNA strand        mismatched base pairs

Figure 1 : Binding of an oligonucleotide to a long template DNA strand, and incorporation into a negative copy of the template, for oligo-directed mutagenesis. X's indicate locations of non-Watson-Crick (mismatched) base pairs.

Footnotes

 

5.

A DNA polymerase is an enzyme that can build a new DNA strand out of the ACGT building blocks, given an old strand to use as a template. 

6.

Single DNA strands have a polarity; one end is named 5' and the other is named 3'. (These names are derived from organic chemistry nomenclature.) Strands always bind to each other in opposite, antiparallel orientations. Polymerization of new strands always proceeds in the 5' to 3' direction. 

7.

Ligase ligates or joins together adjoining ends of DNA strands. 

8.

Meaning the reverse complement.