The development of the PCR revolutionized DNA-based strategies for diagnosis and treatment. It permits the detection, synthesis, and isolation of specific genes and allows one to discriminate among the alleles of a gene differing by as little as one base. It requires only readily available equipment and basic technical skills. A specimen consisting of only minute amounts of material will suffice; in most circumstances, no special preparation of the tissue is necessary. PCR made direct genetic and genomic analyses readily accessible to clinical, epidemiologic, and forensic laboratories. This single advance fueled quantum increases in the use of direct gene analysis for diagnosis of human diseases. Indeed, PCR analysis combined with direct DNA sequencing technologies have largely supplanted older strategies, such as restriction enzyme mapping and DNA/RNA blotting strategies for many research and diagnostic applications, although these older methods remain useful for some niche applications. PCR coupled with now-routinely available gene cloning methodologies allows one to synthesize in microgram quantities naturally occur ring or engineered genes at will. These can then readily be inserted into cells, tissues, or organisms where they will be expressed and their physiologic or pathologic effects investigated. Similarly, industrial scale production of novel therapeutics based on the PCR-designed DNA itself or its expressed RNA or protein products is now routine. Hematopoietic growth factors and monoclonal antibody therapeutics are just two examples of widely used hematologic therapies that depended on these strategies.

PCR is based on the prerequisites for copying an existing DNA strand by DNA polymerase: an existing denatured strand of DNA to be used as the template and primers. Primers are short oligo nucleotides, 12 to 100 bases in length, having a base sequence complementary to the desired region of the existing DNA strand. Oligonucleotide primers are now easily designed and produced using biochemical techniques developed in the 1970s and 1980s. The primer allows the polymerase to “know” where to begin copying. If the base sequence of the DNA of the gene under study is known (see DNA sequencing), two synthetic oligonucleotides complementary to sequences flanking the region of interest can be pre pared. If these are the only oligonucleotides present in the reaction mixture, then the DNA polymerase can copy only daughter strands of DNA downstream from those oligonucleotides. In other words, it can copy only that gene. Recall that DNA is double stranded, that the strands are held together by the rules of Watson-Crick base pairing, and that they are aligned in antiparallel fashion. This implies that the effect of incorporation of both oligonucleotides into the reaction mix will be to synthesize two daughter strands of DNA, one originating upstream of the gene and the other originating downstream. The net effect is synthesis of only the DNA between the two primers, thus doubling only the DNA containing the region of interest. If the DNA is now heat denatured and then cooled again, allowing hybridization of the daughter strands to the primers, and the polymerization is repeated, then the region of DNA through the gene of interest is doubled again. Thus two cycles of denaturation, annealing, and elongation result in a selective quadrupling of the gene of interest. The cycle can be repeated 30 to 50 times, resulting in a selective and geometric amplification of the sequence of interest to the order of 2³⁰ to 2⁵⁰ times. The result is a millionfold or higher selective amplification of the gene of interest, yielding microgram quantities of that DNA sequence.

PCR achieved practical utility when DNA polymerases from thermophilic bacteria were discovered; when synthetic oligonucleotides of any desired sequence could be produced efficiently, reproducibly, and cheaply by automated instrumentation; and when DNA thermocycling machines were developed. Thermophilic bacteria live in hot springs and other exceedingly warm environments, and their DNA polymerases can tolerate 100°C (212°F) incubations without substantial loss of activity. The advantage of these thermostable polymerases is that they retain activity in a reaction mix that is repeatedly heated to the high temperature needed to denature the DNA strands into the single-stranded form. Microprocessor-driven DNA thermocycler machines can be programmed to increase temperatures to 95°C to 100°C (203°F to 212°F) (denaturation), to cool the mix to 50°C (122°F) rapidly (a temperature that favors oligonucleotide annealing), and then to raise the temperature to 70°C to 75°C (158°F to 167°F) (the temperature for optimal activity of the thermophilic DNA polymerases). In a reaction containing the test specimen, the thermophilic polymerase, a sufficient supply of primers to support the amplification, and the chemical components needed to sustain the multiple rounds of copying (e.g., nucleotide triphosphate precursors, reaction buffer, an adenosine triphosphate [ATP]-generating system to support the endothermic polymerase reaction), the thermocycler can conduct many cycles of denaturation, annealing, and polymerization in a completely automated fashion. The gene of interest can thus be amplified more than a millionfold in a matter of a few hours. The DNA product is readily identified and isolated by routine agarose gel electrophoresis. The DNA can then be analyzed by restriction endonuclease, digestion, hybridization to specific probes, sequencing, further amplification by cloning, and so forth.

Reverse transcriptases (RNA-dependent DNA polymerases) derived from retroviruses greatly extend the utility of PCR. By copying all the RNAs into their cDNAs, reverse transcriptase allows RNA sequences in a specimen to be amplified much like DNA sequences. This procedure, called reverse transcription (RT)-PCR, inserts a reverse transcriptase step into the beginning of the procedure, which then proceeds exactly like PCR. RT-PCR permits one to amplify all of the mRNAs expressed in a cell for high-throughput nucleotide sequence analysis, to detect just one or a few mRNAs to analyze their expression patterns, or to clone them (see later) to isolate their encoding genes.