Polymerase chain reaction

INTRODUCTION — The two most important principles underlying the polymerase chain reaction (PCR) are: Complementarity-driven binding of DNA to form a duplex Template-driven, semi-conservative synthesis of new DNA by DNA polymerases
These two principles are discussed separately in detail. (See "Overview of molecular biology").
To complete the theoretical background to understand PCR, a discussion relating to the requirements that specify a unique sequence in the genome, and a review of thermostable DNA polymerases are required. These features as well as a discussion of some research applications of PCR are presented here. An introduction to the clinical applications of PCR can be found elsewhere within UpToDate. (See "Cytogenetic and molecular genetic diagnostic tools").
UNIQUE GENOMIC SEQUENCES — Relatively short DNA sequences suffice to specify a unique sequence in the genome. As an example, assume (although it is not strictly true) that all four bases (adenine [A], thymine [T], guanine [G], and cytosine [C]) are equally frequent in the approximately 3 X 10(9) residues of the human genome. It is possible to calculate the expected frequency of a sequence of N bases having a specified sequence using simple probability theory; that frequency is given by the following formula:
(3 X 10(9))[1/4(N)]
The expression 1/4(N) expresses the chance that the sequence of an N-residue oligonucleotide matches a given sequence. The constant 3 X 10(9) is simply the size of the human genome in bases. The following are examples of this calculation for four different oligonucleotide lengths: A specific oligonucleotide that is 10 nucleotides long occurs approximately 3000 times in the human genome. An oligonucleotide that is 15 nucleotides long occurs approximately 3 times. Specific oligonucleotides that are 20 and 25 nucleotides in length occur at an approximate frequency of 0.003 and 0.000003, respectively.
Although the precise frequencies vary from these examples because of the simplified assumptions regarding base abundance and the independence of sequence at a specific residue from sequence at neighboring residues, these rough estimates demonstrate that a specific 20-residue sequence is very likely to be unique in the human genome.
THERMOSTABLE DNA POLYMERASES AND SYNTHETIC OLIGONUCLEOTIDES — The existence of thermostable DNA polymerases, purified or cloned from microorganisms living in hot springs, is a necessary component of PCR. These polymerases can withstand heating to 95 degrees C with minimal loss of activity and function optimally near 70 degrees C. The ability to readily and inexpensively synthesize specific oligonucleotides of 20 to 30 residues is the other technical underpinning of PCR.
PROCESS OF PCR — A reaction mixture containing a large excess of a pair of oligonucleotide primers designed to match either end of the target sequence, a substrate DNA, free deoxynucleotide triphosphates, and a thermostable DNA polymerase is assembled. The mixture is heated to 95 degrees C to allow the double stranded substrate DNA to denature into single strands. The mixture is then cooled to a temperature just below the predicted denaturation temperature of the primers, which will then anneal to the single-stranded substrate DNA and prime new DNA synthesis by the included DNA polymerase. The temperature is then raised to the optimal temperature for the polymerase to allow chain elongation to proceed long enough for synthesis to extend past the opposite primer's complementary sequence. The mixture is then heated to 95 degrees C again in order to once again separate all the DNA to single strands, and the entire sequence of temperature cycling is repeated. Now, the number of potential targets for primer annealing has been doubled, as both the original substrate DNA and the newly synthesized strands are available. In general, between 25 and 40 such cycles of denaturation, annealing, and elongation are performed, resulting in exponential amplification of the target sequence.
Early experiments were performed by physically moving reactions among three water baths preset to the desired temperatures. Currently, a variety of automated thermal cyclers perform the desired temperature regulation with minimal operator hands-on time. All the components of the reactions are readily available commercially, and computer programs that facilitate primer design and calculate annealing temperatures are freely distributed on-line [1-3].
OVERVIEW OF RESEARCH APPLICATIONS — PCR's impact on biomedical research has been immense. This technology allows large quantities of rare sequences to be synthesized, cloned, and analyzed with high reliability and minimum effort. The award of the 1993 Nobel Prize in Chemistry to Kary B. Mullis for inventing the technique recognized the importance of PCR-based methods. A few examples of common research applications of PCR are briefly described below.
PCR is a central tool in genomics and genetics. Relatively early in the human genome project, it was recognized that PCR technology permits more extensive and easier sharing of reagents than had been possible previously. To share a clone, for example, investigators need only specify a pair of primer sequences, the size of the expected product, and buffer conditions for its successful amplification. Any laboratory receiving this information subsequently has the capability of amplifying the sequence from genomic DNA and cloning it into a suitable vector [4]. This is obviously much easier than exchanging actual specimens or cultures.
Amplification of genomic DNA — Two examples of amplification of genomic DNA by PCR will briefly be presented. These are genotyping at microsatellite markers and detection of rare sequences. Both of these examples are fairly common research applications and are likely to be adapted to the clinical setting in the near future.
Microsatellite genotyping — To provide genetic markers, primer pairs flanking short, repetitive sequences called microsatellites were chosen. These were found to be present in variable numbers of copies at a population level, but to display Mendelian inheritance within families. Copy number of the repetitive sequence within the amplified product can therefore define various alleles, while the unique primers define a specific genomic location. This is discussed more extensively in the topic review on repetitive DNA sequences. (See "Repetitive DNA").
Detection of rare sequences — PCR technology has also allowed detection of rare DNA sequences in a population of DNA molecules. This application is particularly prominent in searching for DNA rearrangements in the setting of neoplasia. An example is the discovery that a Herpes simplex-related virus is involved in the pathogenesis of Kaposi's sarcoma [5-7]. Representation difference analysis, a PCR-based method for preferentially amplifying sequences present in one of a pair of sources of substrate DNA [8], has allowed investigators to determine that tumor tissue (but not surrounding normal tissue from the same individuals) harbored integrated viral DNA. Thereafter, primers were designed to allow direct amplification of the viral sequences. (See "Cytogenetic and molecular genetic diagnostic tools" and see "Diagnosis and antiviral therapy of human herpesvirus 8 infection").
Amplification of RNA — PCR can be applied to messenger RNA (mRNA) by addition of a reverse transcription (RT) step prior to amplification. RT-PCR allows detection of rare messages whose abundance is below the detection threshold for Northern blot analysis. Moreover, it is easier and faster to perform. Two examples are expression profiles and RNA virus infection.
Expression profiles — A vast area of current research assesses changes in patterns of gene expression in response to various perturbations. RT-PCR allows qualitative, semi-quantitative, or quantitative measurement of mRNA levels. Qualitative and semi-quantitative assays can now be performed on microarrays, permitting thousands of genes to be studied simultaneously (reviewed by [9-12]).
RNA virus infection — PCR of RNA isolated from blood has become a standard tool in monitoring the viral load in HIV-infected patients (reviewed by [13,14]). This strategy is being extended to other RNA viruses causing chronic infection, such as hepatitis C [15,16]. (See "Techniques and interpretation of HIV-1 RNA quantitation" and see "Diagnostic approach to hepatitis C virus infection").
CRITICAL EVALUATION OF DATA — The ease of performing PCR has led to wide dissemination of the methodology. The rigor with which work is done varies enormously. When reading medical literature including PCR data, one must evaluate the work critically since publication is not a guarantee of high-quality work.
Every PCR experiment requires a minimum of two technical controls, in addition to biological controls specific to the research question being addressed. The technical controls should demonstrate that amplification occurs when it should (positive control) and that it does not occur when it should not (negative control).
In RT-PCR, it is also important to distinguish semi-quantitative from truly quantitative designs. Because substrate concentrations may become limiting in later rounds of amplification, comparing intensity of electrophoretic bands in most circumstances is only semi-quantitative [17,18]. True quantitation generally entails either inclusion of an internal standard that is coamplified competitively with the target or use of special methodology, such as real time PCR [19-22]. (See "Cytogenetic and molecular genetic diagnostic tools")