to BioTechniques free email alert service to receive content updates.
Gene Splicing by Overlap Extension: Tailor-Made Genes Using the Polymerase Chain Reaction
Robert M. Horton, Zeling Cai, Steffan N. Ho, and Larry R. Pease
Mayo Clinic
BioTechniques, Vol. 54, No. 3, March 2013, pp. 129–133
Full Text (PDF)

Gene Splicing by Overlap Extension or “gene SOEing” is a PCR-based method of recombining DNA sequences without reliance on restriction sites and of directly generating mutated DNA fragments in vitro. By modifying the sequences incorporated into the 5′-ends of the primers, any pair of polymerase chain reaction products can be made to share a common sequence at one end. Under polymerase chain reaction conditions, the common sequence allows strands from two different fragments to hybridize to one another, forming an overlap. Extension of this overlap by DNA polymerase yields a recombinant molecule. This powerful and technically simple approach offers many advantages over conventional approaches for manipulating gene sequences.

The polymerase chain reaction (PCR) (10) has quickly become a fundamental analytical tool in molecular biology. Recently, several laboratories, including our own, have begun to use this technology as a synthetic tool for recombining DNA sequences without relying on restriction sites. Here we describe the technique and its uses and review briefly some related PCR applications.


In PCR, primer extension by DNA polymerase is used to make a copy of a DNA strand which can then serve as a template for extension from a second primer in the opposite orientation. Multiple rounds of this process lead to exponential accumulation of the sequence of interest. The result of a PCR is a DNA segment of defined length which has incorporated synthetic oligonucleotide primers into its ends. The 3′-ends of these primers must match the sequence of the template gene well enough to act as primers for DNA polymerase, but the 5′-ends can include sequences unrelated to the template gene. This capability, called “mispriming” by Mullis et al. (10), has been used to perform site-directed mutagenesis (2, 8) and to add sequences to the end of a PCR generated fragment (11, 14). The limitation of this technique is that the mutation or added sequence must be in the primer, and thus must be within the length of an oligonucleotide from the end of the PCR fragment. If the fragment is to be cloned, restriction sites would normally have to be included in the primers as well. This means that this method can only be used to make changes at positions close enough to the restriction sites used for cloning to be included it the same oligonucleotide.

Overlap Extension

Higuchi et al. (5), inspired by Mullis et al. (11), were the first to report the use of a technique to introduce mutations into the center of a PCR fragment making PCR mutagenesis much more flexible. Ho et al. (6), who independently developed the technique and refer to it as “overlap extension,” also demonstrated that the error frequency intrinsic to this method is sufficiently low to make it practical for widespread use. A modification of this method (7) allows segments from two different genes to be recombined or “spliced” together by overlap extension, a process we refer to as “gene Splicing by Overlap Extension” (SOE) or “gene SOEing.”

The mechanism is illustrated in Figure 1. PCR is used to generate two fragments (AB and CD) which have their ends modified by mispriming so that they share a region of homology. When these two fragments are mixed, denatured and reannealed, the 3′-end of the top strand of fragment AB anneals onto the 3′-end of the bottom strand of fragment CD, and this overlap can be extended to form the recombinant product. The overlap region is determined by primers ‘b’ and ‘c’ and can contain any sequence, as long as the oligomers are complementary. This region is where base changes are incorporated when the technique is used for site-directed mutagenesis. Alternatively, the overlap can be designed to make a “neat” joint between two fragments, with no new sequences included, as in the present report.

In this report we describe the construction of a recombinant gene encoding a chimeric protein in which parts of a class I major histocompatibility complex (MHC) antigen are replaced by corresponding portions of a distantly related class II MHC antigen.

Materials and methods


Oligonucleotide primers were synthesized on an Applied Biosystems Model 380A DNA synthesizer (Foster City, CA) and desalted on a Sephadex G-50 column (Pharmacia LKB Biotechnology, Piscataway, NJ).


The class I sequences were derived from a plasmid containing Kb(13), and the class II sequences were derived from plasmids containing Aαk(1) and Aβk(3).

Primer Design

The sequences of the eight primers used for this recombination are given in Table 1. Primers ‘a’ and ‘h’ are the flanking or “outside” primers, which serve to PCR amplify the final recombinant product. They do not contribute to the sequences added at the overlapping ends. Oligomers ‘b’ and ‘c,’ ‘d’ and ‘e,'and ‘f’ and ‘g’ are the SOEing primers. The members of each pair are related because bases have been added to their 5′-ends to make them complementary to one another. In each case, the overlap region between the primers, and the priming region by which each primer recognizes its template, was designed to have an estimated Td of approximately 50°C according to the formula Td= 4 (C + G) + 2(A + T) in degrees Celsius (15). In practice, we have found that simply making these regions 15 to 16 nucleotides long generally works well. We have not made a careful examination of the minimum length of the oligomers.

  1    2    3