Site-directed mutagenesis is a method used to alter the nucleotide sequence of cloned DNA at a predefined position or site. This method is one form of in vitro mutagenesis, because it involves a series of biochemical steps using purified reagents and is performed on cloned DNA in a small test tube. Site-directed mutagenesis can be used to change only a single base pair, to change a few base pairs, or to create more extensive sequence changes, such as deletions and insertions. The ability to create site-directed mutations in cloned DNA largely became possible with the advent of methods to synthesize oligonucleotides, short segments of DNA, to create the desired DNA changes (see DNA Synthesis). Although methods using chemical modification of DNA can create site-directed mutations in cloned DNA (see Mutagenesis ), site-directed mutagenesis is most easily and precisely accomplished using an oligonucleotide; hence, the method is often termed oligonucleotide-directed mutagenesis. This entry describes the basic principles of oligonucleotide-based site-directed mutagenesis and highlights a number of procedures currently in use.
1. Site-Directed Mutagenesis is a Powerful Tool for Studying Nucleic Acids and Proteins
The ability to alter cloned DNA at will provided molecular biologists with a powerful tool with which to study the role of cloned DNA in cellular processes. Prior to the development of site-directed mutagenesis, biologists’ only source of DNA variants was from rare spontaneous mutations that occurred randomly in vivo or from imprecise methods for in vitro chemical mutagenesis of cloned DNA. In contrast, oligonucleotide-based site-directed mutagenesis changes the sequence of a DNA fragment precisely as designed by the molecular biologist. Cloned genes can be studied in a biological system by comparing the function of the normal gene to variants containing site-directed mutations.
This revolutionary technique is not limited to the study of genes. Because the flow of genetic information is normally DNA ^ RNA ^ protein (according to the central dogma), site-directed mutagenesis of a cloned DNA fragment can also be used to study RNA and proteins. For example, site-directed mutagenesis has been used to uncover nucleotides in messenger RNA precursors that direct the RNA splicing of introns (1), to characterize amino-acid residues critical for protein-protein interactions (2), and to probe the catalytic mechanism of enzymes (3). There are many additional examples where site-directed mutagenesis has helped to solve basic problems of biology, and it has also been applied in the biotechnology industry to generate novel therapeutics.
Since the first report of oligonucleotide-based site-directed mutagenesis (4) by M. Smith and colleagues in 1978, site-directed mutagenesis has developed into a number of related methodologies that are now part of every molecular biologist’s toolbox. Kits containing the reagents and protocols needed to conduct site-directed mutagenesis are commercially available and make the technique relatively straightforward to perform. In 1993, the method was recognized as being truly revolutionary by the award of the Nobel Prize for Chemistry to Smith for his "fundamental contributions to the establishment of oligonucleotide-based, site-directed mutagenesis and its development for protein studies" (5).
2. Two General Methods for Oligonucleotide-based Site-directed Mutagenesis
Current procedures for oligonucleotide-based site-directed mutagenesis can be grouped into two categories: (1) cassette mutagenesis and (2) enzymatic extension of a mutagenic oligonucleotide annealed to a DNA template. Cassette mutagenesis is, in practice, the simpler of the two approaches. As described in the next section, cassette mutagenesis is a derivative of total gene synthesis (6) in which a gene is constructed by ligating a series of synthetic oligonucleotides.
Site-directed mutagenesis by enzymatic extension of mutagenic oligonucleotides stemmed from research in the 1960s and 1970s on DNA polymerization in vitro, the genetics of bacteriophage wX174, and the preparation of synthetic oligonucleotides. Researchers were studying the ability of the enzyme DNA polymerase I from Escherichia coli to synthesize DNA in a test tube using an oligonucleotide primer hybridized to a single-stranded DNA template from wX174 bacteriophage (7). These experiments demonstrated that a single-stranded DNA template could be converted to a double-stranded molecule in vitro . Importantly, the DNA polymerase needed a short oligonucleotide primer to initiate DNA synthesis. Additional experiments on wX174 studied the ability of restriction fragments from wild-type wX174 annealed to a mutant form of wX174 to create wild-type phage by a process called marker rescue (8). These experiments, and research on oligonucleotide synthesis, led to the first site-directed mutagenesis experiment, where an oligonucleotide with only 12 nucleotides was used to change a single G to A in the wX174 DNA. Since the initial experiments on wX174 (4, 9), site-directed mutagenesis protocols have been developed for DNA fragments cloned into a variety of vectors. These methods can be subdivided further, depending on the particular DNA polymerase used. Early methods used DNA polymerases derived from E. coli or bacteriophage T4 or T7. Recent procedures have been developed for DNA polymerases from thermostable bacteria and the polymerase chain reaction (PCR). The details of these methodologies are discussed below.
2.1. Synthetic DNA Cassettes
In cassette mutagenesis (see Fig. 1), a restriction fragment from the cloned DNA of interest is replaced by another restriction fragment containing the desired nucleotide changes (10-12). The process begins by digesting the cloned DNA with one or two restriction enzymes to release a DNA fragment containing the (wild-type) sequence to be changed. The new (mutant) sequence is constructed from two synthetic, complementary single-stranded oligonucleotides, which are first mixed together in a buffered solution so they hybridize, thereby forming a duplex DNA molecule, but with cohesive, sticky ends to hybridize with restriction sites of the vector. To position the synthetic fragment back into the vector, the ends of the synthetic DNA must be the same as that in the wild-type fragment. The remainder of the cloned DNA in the vector is mixed with the synthetic fragment, and the two are ligated into place by the action of the enzyme DNA Ligasefrom bacteriophage T4. Following the joining of the two DNA molecules, DNA from the ligation reaction is introduced into competent E. coli cells, and recombinant molecules containing the desired mutation are selected.
Figure 1. Cassette mutagenesis. A plasmid containing a clone of your favorite gene (YFG; black segment) is cleaved with two restriction enzymes, eg, XbaI and BglII, each of which has only one restriction site in the entire plasmid. The reaction mixture is separated by agarose gel electrophoresis, and the larger fragment is purified from the gel. Two single-stranded oligonucleotides are synthesized by automated DNA synthesis. The sequences of the oligonucleotides are complementary to each other and differ from the wild-type sequence at only a single position (black stripe) containing the desired changes. The oligonucleotides are mixed in a solution that promotes hybridization of the two strands by virtue of their complementary sequences. The ends of the duplex fragment are single-stranded, cohesive, sticky ends that join with XbaI and BglII sites. The DNA cassette is mixed with the isolated fragment, and the two molecules are covalently joined by the action of T4 DNA ligase. The ligated DNA is transformed into E. coli, and drug-resistant colonies are selected. Plasmid DNA is prepared from individual bacterial colonies. Since the two linear fragments themselves cannot transform E. coli, all colonies contain plasmids with the mutant sequence.
Cassette mutagenesis generally results in nearly 100% of the resulting clones having the desired new (mutant) sequence. The difficulty with this method is that unique, conveniently spaced restriction endonuclease recognition sites to excise a small cassette in the desired place are often not present. In this case, one can turn to enzymatic extension methods.
2.2. Enzymatic Extension of a Mutagenic Oligonucleotide
The principle and end result of site-directed mutagenesis by oligonucleotide extension are similar to those of cassette mutagenesis: An oligonucleotide encoding the desired new DNA sequence is inserted into a cloned DNA fragment. In the basic method (Fig. 2), a short oligonucleotide, typically 20 to 40 nucleotides in length, is hybridized to its complementary sequence in a circular, single-stranded, wild-type DNA template. The sequence of the oligonucleotide is designed so that the new (mutant) sequence is in the middle of the oligonucleotide and flanked by wild-type sequences on the ends. Upon annealing the oligonucleotide to the template DNA, the wild-type sequences within the oligonucleotide are perfectly complementary to the template DNA, but the mutant sequences are mismatched. The wild-type sequences need to be long enough so the oligonucleotide will form a stable duplex with the template DNA. Next, a DNA polymerase and nucleotide precursors (dNTPs) are added to synthesize the remainder of the complementary strand using the mutagenic oligonucleotide as a primer for DNA synthesis. In the past, derivatives of E. coli DNA polymerase were used in site-directed mutagenesis. Recently, DNA polymerases from bacteriophage T4 or T7 have replaced the E. coli enzyme. These DNA polymerases are more processive than the E. coli enzyme (see DNA Polymerase Sliding Clamps) and do not displace or proofread the mutagenic oligonucleotide, thereby improving the frequency of the mutant plasmids obtained.
Figure 2. Site-directed mutagenesis by enzymatic extension of a mutagenic oligonucleotide. A DNA fragment encoding YFG is cloned into a plasmid vector. Single-stranded DNA is prepared by one of several methods. A synthetic oligonucleotide is annealed to the DNA template. The sequence is perfectly complementary to the wild-type sequence of the template, except for the mismatched nucleotides (black dot) that contain the mutant DNA sequence. The remainder of the complementary strand is synthesized by a DNA polymerase using the mutagenic oligonucleotide as a primer. Upon completing the circle, the ends of the newly synthesized DNA strand are joined by DNA ligase, forming a heteroduplex of two covalently closed DNA molecules. One strand contains the mutant sequence, and the other contains the wild-type sequence. The heteroduplex DNA is introduced into competent E. coli cells. Drug-resistant colonies that contain plasmid DNA are selected. The two strands separate by DNA replication in the bacterial cells. Plasmid DNA is isolated from the bacteria and sequenced to identify clones with the desired mutation.
Once synthesis of the complementary DNA strand has proceeded completely around the circular template, the ends of the newly synthesized strand are joined enzymatically by T4 DNA ligase. In this way, a single-stranded DNA molecule is converted into a double-stranded molecule, where all sequences will be wild-type except the region containing the mismatched nucleotides. The resulting double-stranded heteroduplex DNA molecule, composed of one wild-type strand and one mutant strand, is introduced into competent E. coli cells, where the two strands segregate by DNA replication. Theoretically, each colony should contain both wild-type and mutant plasmids; in practice, however, only one sequence is often obtained from a colony due to mismatch repair prior to replication of the two strands. Generally, the frequency of mutant plasmids is lower than that of the wild type. This is most likely due to incomplete DNA replication in vitro. A number of selection or screening methods have been developed to enrich for plasmids with the desired mutation (see below).
3. Modifications and Improvements to Site-directed Mutagenesis Methods
The first oligonucleotide-directed mutagenesis experiments were performed using DNA from the single-stranded bacteriophage wX174. Because wX174 is not a convenient cloning vector, mutagenesis of cloned DNA fragments was performed using plasmids such as pBR322 (13) or vectors derived from the filamentous M13 phage (14-17) and fd (18). The phage vectors offered an easy way to prepare a single-stranded DNA template. However, protocols were also developed for oligonucleotide-directed mutagenesis in double-stranded plasmid vectors (19, 20). In addition, phagemid vectors (21, 22) were developed that normally are double-stranded but can be induced to produce single-stranded phage-like DNA. These vectors facilitated template preparation for site-directed mutagenesis.
The early methods for oligonucleotide-directed mutagenesis resulted in a low frequency of plasmids with the desired mutation. Often, the fraction of mutants was less than 1 out of 100 plasmids. As discussed above, this was probably due to a variety of technical and biological problems. A number of methods were introduced to make it easier to find a plasmid carrying the desired mutant sequence. For some experiments, mutant molecules can be identified if the mutation happens to create or destroy a restriction endonuclease site. In a more general method, the double-stranded heteroduplex molecules were purified by centrifugation (17, 23). A simpler procedure was to make the mutagenic oligonucleotide radioactive and to use it as a hybridization probe (24). Under certain stringent conditions, the probe would hybridize with only mutant molecules and not with wild-type molecules. Other high-frequency mutagenesis methods used two oligonucleotide primers (25, 26); one was the mutagenesis oligonucleotide, and the other was positioned upstream. The second primer helped to complete synthesis of the complementary strand.
A powerful adaptation of the original procedure (27), developed by Kunkel in 1985, yielded mutants with efficiencies greater than 50%. First, the DNA vector was grown in a dur, ung~ strain of E. coli that causes uracil to be incorporated into the DNA template. Single-stranded template was prepared and annealed with the mutagenic oligonucleotide; the complementary strand was synthesized using standard deoxyribonucleotide triphosphates. In this way, the mutant strand contains thymine and the wild-type strand contains uracil. The heteroduplex molecule is introduced into an ung+ E. coli strain, where the wild-type strand is destroyed by uracil deglycosylase (the product of the ung gene) cleavage of sites containing uracil.
Several other methods are available that increase the efficiency of mutagenesis. One method incorporates thiophosphoryl nucleotides into the newly synthesized strand, allowing the wild-type strand to be converted into the mutant sequence in vitro (28). Yet another method uses a second primer that alters a restriction site or corrects a mutation in a drug-resistance marker (29, 30). In this way, plasmids containing the desired mutation can be enriched by drug selection or restriction enzyme digestion. By most of these improved methods, the fraction of mutant molecules obtained was often greater than 50%. Once these highly efficient methods were established, mutants were identified simply by randomly picking one or two plasmids and subjecting them to DNA sequencing. Finally, the ability to perform multiple mutagenesis experiments in parallel has been addressed by the development of solid-phase mutagenesis methods (31). By this approach, the creation of site-directed mutants can be automated.
