The manipulation of genetic material through the selection of organisms with desired heritable traits or characteristics probably dates back to the earliest forms of human agricultural activity. The most ancient form of genetic selection requires only that the recognition of a naturally occurring variant possessing the desired trait be accompanied by the selective breeding of the variant to establish many individuals carrying the desired characteristic. The selection of grains that retained rather than dropped mature seeds, saving seeds from particularly large fruits and vegetables, and the breeding of livestock for increased meat, wool, or milk production are all examples of genetic engineering in which simple selective techniques were used to manipulate genetic characteristics with only a minimal understanding of the nature of genetic information. Agronomic activities (husbandry) allowed man to impose artificial selections for desired characteristics not necessarily advantageous to an organism in a natural setting. The key to breeding of modified strains, as in natural selection, is that a selectable phenotype must be linked with a heritable genotype; characteristics that are acquired, rather than inherited, cannot be bred. This lowtechnology genetic manipulation, although slow and painstaking, has nevertheless been historically very effective in the development of desired genetic characteristics since long before humans had an understanding of the chemical nature of the machinery of inheritance. Virtually all agronomically important fruits, grains, vegetables, and livestock are the products of this form of genetic engineering.
One of the first scientific uses of the term “genetic engineering” was, in fact, by Hansche et al. in the October 1971 issue of Diamond Walnut News in describing how developments had turned “varietal improvement into a problem of biological engineering and the plant breeder into a genetic engineer” (1). Scientific progress in the area of understanding gene structure and function was so rapid that, less than three years later, at a National Academy of Sciences workshop (Genetic Improvement of Seed Proteins), the concept of introducing specific traits via the isolation and manipulation of DNA was already under discussion (2). In a very short period of time, “genetic engineering” had changed from a term characterizing the selection of desired traits to a concept describing the isolation, modification, and introduction of genes encoding the desired traits. The term suggests that genes can be conceived of as molecular machines capable of being designed to modify their performance. This concept of genetic engineering as involving the alteration of specific molecules was made possible by a virtual explosion in the understanding of gene structure and function. Although the broader use of the term describes the manipulation of organisms at the genomic level, in this volume we focus on genetic engineering in the narrow sense of the manipulation of specific genes.
In the latter half of the twentieth century, the elucidation of deoxyribonucleic acid (DNA) as the macromolecule responsible for long-term storage of genetic information (3-5) and ribonucleic acid (RNA) as the molecule involved in conveying genetic information to the protein synthesis machinery of living cells (6) resulted in the development of three successive approaches to the specific manipulation of DNA to achieve the desired genetic engineering goals. These approaches can be thought of as conventional DNA genetics, restriction enzyme-based recombinant DNA methodology, and PCR technology. Each of these technologies uses a different means to obtain a DNA molecule with specific physical and genetic characteristics. Although the three approaches are in many ways interrelated, each has specific strengths and limitations that make its application most suited to specific types of problems. A brief overview of these three generations of genetic engineering techniques may help develop an appreciation of the tremendous power of PCR-based technologies in the specific manipulation of DNA molecules.
The independent assortment of allelic pairs that accompanies sexual reproduction was established early in the twentieth century as a fundamental tool in the characterization of gene structure and function (7-9). Following the identification of DNA as the genetic material (3-5), it became increasingly obvious that manipulation of genetic traits required manipulation and modification of DNA molecules. Transduction, the ability of a bacteriophage or virus to transfer traits to a recipient, and conjugation, the transfer of traits from a donor to a recipient bacterium, were both determined to involve transfer of DNA (rev. in 10), as was the transfer of drug resistance among bacteria (rev. in 11).
Variants or mutants, whether naturally occurring or generated by chemicals, viruses, or transposons, provided a way to mark genes. Selection for revertants after gene transfer provided a way to identify functional counterparts of the mutated genes. The relative power and utility of these natural DNA engineering mechanisms can be demonstrated by the observation that, prior to the advent of either recombinant DNA or PCR technologies, fairly extensive genetic maps had been established for the genomes of bacteriophage lambda, E. coli, D. melanogaster, and some yeasts and fungi. In addition, the genetic organization and some genetic control mechanisms of both the lac operon and bacteriophage lambda had been substantially elucidated.
The limitations of the first generation DNA engineering technology, however, are tremendous: DNA manipulations are limited to a genetic system in which a naturally occuring DNA transfer mechanism has been found, the relative efficiency of obtaining a specific DNA construct can be vanishingly small due to the low prevalence of the gene of interest in the total genome, and the ability to create sufficient numbers of genetically identical individuals may be limited.
The importance of the generation of genetically identical individuals cannot be overstated. An organism that is capable of asexual reproduction, such as bacteria, yeasts, and some plants, can easily be manipulated to give rise to a colony of isogenic individuals or clones. Genetic inbreeding must be used in other species, particularly in higher eukaryotes, to establish colonies or groups of essentially genetically identical organisms. Since a single gene represents such a small fraction of the total genome of an organism, the cloning or inbreeding process provides a means to amplify the total DNA of an organism, thereby increasing the amount of the gene of interest.
Though progress was made in organisms such as the genetically well-characterized fruit fly D. melanogaster and the extensively inbred lines of laboratory mouse Mus musculus, the lack of available genetically defined strains made many DNA engineering experiments simply not feasible using first-generation technology.
Many of the solutions to the principle obstacles confronting first-generation DNA engineering technology were developed during the application of conventional DNA genetics to various bacterial systems. One of the keystones of second-generation DNA technology was the intriguing observation that a bacterial strain could be relatively resistant to infection by a bacteriophage stock grown on a different but closely related bacterial strain. The infrequent infection event that occurred, however, could give rise to a bacteriophage stock adapted to efficient infection of the second bacterial strain, but no longer able to efficiently infect the original host (12). This observation ultimately resulted in the discovery and characterization of restriction enzymes, naturally occurring enzymes that recognize and cleave DNA in a site-specific manner.
Because these enzymes recognize and cleave at specific sequences along a DNA molecule (13), they allowed conversion of a mixture of randomly sheared chromosomal DNA fragments to a set of DNA fragments with sizes determined by the location of restriction cleavage sites within the DNA molecules. For small genomes, such as those of bacteriophages, plasmids, and viruses, this innovation alone was sufficient to allow the physical characterization of DNA molecules and the correlation of certain genetic traits with specific DNA fragments. However, analysis of DNA restriction fragments alone was not sufficiently powerful to allow efficient analysis of larger genomes.
The second critical innovation that allowed the evolution of second generation DNA technology was the concept of using small, naturally occurring replicons as “molecular vehicles” (2), or vectors, to allow propagation and biological amplification of specific DNA fragments (14, 15). Restriction digestion of DNA followed by mixture with a vector DNA molecule allowed the annealing of the cohesive restriction termini to cause the formation of new, recombinant DNA molecules. When the nicks in the annealed termini were sealed with DNA ligase, either in vitro prior to or in vivo after insertion of the recombinants into a host, each resulting construct allowed whatever DNA fragment happened to have been ligated to the vector to be isolated by cloning the host organism. This innovation circumvented the need to amplify the entire genome of an organism to obtain workable amounts of a specific gene: an individual gene could be amplified by growing a recombinant DNA molecule in a bacterial clone. This simple and useful form of gene engineering provided the foundation for early transgenic technology that used extrachromosomal vectors for the overproduction of recombinant proteins in bacteria.
The ability to specifically cleave, biologically amplify, and isolate specific DNA fragments greatly enhanced the efficiency of construction and isolation of a desired DNA molecule and gave rise to the recombinant DNA methodology of the seventies and eighties. It provided the means to take molecular machines apart and rebuild them to suit human needs. When combined with improved screening technologies, DNA sequence analysis, and DNA synthesis, the second generation DNA technology gave rise to an avalanche of information concerning gene structure and function. Great databases were established for gene sequences, genomic structure information, and even sequences of genomes.
In spite of the tremendous power afforded by this technology, recombinant DNA methodology still suffered from key limitations when applied to many specific genetic engineering goals. The ability to make desired recombinant molecules was often dependent on the fortuitous occurrence of restriction cleavage sites. Even in cases where construction of a specific construct was technically feasible, finding the desired molecule amongst a background of undesired molecules might require exhaustive screening. In addition, the need to amplify the desired DNA fragment on a vector in a host at times resulted in the generation of deletions, insertions, or rearrangements in the cloned DNA fragment. The presence of some recombinants has proven to be detrimental to survival of the host, and a specific host will not necessarily propagate all recombinants. Once the desired DNA fragment has been found, the same exhaustive screening must be repeated if the same gene is required from a different organism. Thus, the requirement for a biological host, one of the intrinsic components of the second generation DNA technology, can become a significant obstacle to application of recombinant DNA cloning technology to certain genetic engineering problems.
Just as the development of first-generation DNA technology gave rise to the second-generation recombinant DNA technology, the availability of cloned, isolated DNA fragments of defined nucleotide sequence was of fundamental importance to the development of the third generation of DNA engineering: the in vitro manipulation of DNA by means of the polymerase chain reaction (PCR). The invention of PCR (16-18) led to another fundamental change in the way genetic material can be analyzed and exploited. Whereas second-generation DNA engineering approaches are to a great extent dependent on the availability of enzyme recognition sites within the DNA of interest and on the amplification of the desired DNA construct by cloning into a host organism, third-generation approaches utilize the specificity of chemically synthesized oligonucleotides and the highly efficient in vitro amplification of DNA to enhance the ability to make desired engineered DNA molecules (for review of PCR concepts see Appendix 1).
PCR-based sequence manipulation encompasses several applications that, in the second-generation technology, were typically performed through a succession of separate methodologies. Isolation, amplification, and identification of specific gene fragments, as well as site-directed and random mutagenesis and recombination, can be achieved by simple methodologies that can often be completed in a matter of hours.
Many of these techniques are applications of the principle of overlap extension (19-23), a technique in which the 3' terminus of an amplified sequence segment is used as a PCR “megaprimer” in a subsequent amplification reaction (24). This technique provides a method to perform site-directed mutagenesis on PCR products at an arbitrary distance from the ends of the molecule, and, if the megaprimer and the subsequent template are amplified from different genes, then the product is a recombinant molecule that has been engineered, isolated, and amplified all by the same process.
In fact, PCR-based DNA technology can be used to perform simultaneous mutation and recombination, such as the engineering of a single cysteine residue into an immunoglobulin heavy chain (Chapter 10). The application of PCR techniques has blurred the distinctions among mutagenesis, recombination, and de novo synthesis of genes. The product of PCR-based manipulations is really a mosaic in which sequences derived from natural sources are connected by sequences derived from the synthetic oligonucleotide primers used to direct the amplification; essentially any desired gene sequence can be constructed by combining natural, mutant, and synthetic regions.
Other approaches to PCR-based DNA sequence manipulation do not rely on overlap extension. These include ligation of engineered overlaps (25, 26) and engineering of homologous ends that can undergo homologous recombination in vivo (27, 28). Many PCR applications, such as using PCR primers containing type IIS restriction sites to generate sticky ends of arbitrary sequence (i.e., not including the restriction site itself, which is removed from the molecule when the strands are cut) to direct the ligation of PCR products (Chapter 5), bridge both second- and third-generation technologies. Other approaches that take advantage of both technologies include the use of mispriming to introduce or eliminate restriction sites (29), generation of cohesive ends by removal of chemically distinct primers (30), and generation of cohesive ends by depriving a proofreading polymerase of certain bases (31, 32).
Many significant developments in PCR technology have had tremendous impact on the versatility of PCR for sequence manipulation. Problems related to the errors that may be introduced in the amplification process, for example, were significantly reduced by the description of high-fidelity reaction conditions for the use of Taq polymerase (33) and by the isolation and production of thermostable proofreading polymerases with lower error rates (34-36). Development of conditions that facilitate the amplification of long segments (37, 38) was also a key to increasing the general applicability of PCR-mediated DNA manipulation. As with the second-generation recombinant DNA technology, the development of third-generation PCR-based technology has been the cumulative result of the efforts of many researchers.
PCR-based approaches are not only extremely powerful, they are extremely democratic, in the sense that a broad audience can use them because they are quick and straightforward. Whereas second-generation DNA cloning technology is dependent on a freezer full of heat-labile proteins including restriction enzymes, ligase, phosphatase, and various RNA and DNA polymerases, PCR can provide tremendous logistical simplification in a molecular biology lab, since so many things can be done with a single technology involving essentially a single enzyme. With failure of a molecular biology experiment, the enzymes involved are immediately suspect. Many of these valuable and essential tools can be ruined by simply being left at room temperature for a few minutes, and the loss of activity is not obvious until the failure of the experiment, perhaps days later. Thermostable DNA polymerases, on the other hand, are typically extremely stable and are generally unaffected by accidental incubation for a few hours at room temperature. The rapidity of PCR amplification also greatly reduces the time necessary to evaluate the success or failure of a fragment manipulation. PCR often eliminates the need for cloning and biological amplification of the desired product, thereby obviating the problems that can be associated with culture maintenance and propagation of biological hosts. In vitro expression of engineered gene products can speed functional analysis (39). The obvious benefits of PCR technology have contributed to the rapid proliferation of these methods in fields involving both basic and applied research.
Our division of the history of DNA technology into three generations is admittedly rhetorical. In reality, these techniques complement and build upon one another and form a tremendously powerful body of technology. Many of the DNA fragment engineering approaches currently in use are mixed, using both second-generation (i.e., restriction enzymes, plasmid cloning, etc.) and third generation (PCR-based) techniques. The ultimate goal of other genetic engineering projects is the introduction of the engineered DNA fragment into a biological organism, where the fragment will subsequently be manipulated by conventional first-generation techniques. Genome characterization projects made possible through the use of both cloning technology and PCR methods are proving to be of tremendous benefit to plant and animal breeders utilizing conventional genetics. The availability of gene sequences established by second-generation methods can be crucial to the design of primers that allow the application of third-generation PCR methods to heterologous organisms. Analytical applications of PCR can further accelerate genome mapping efforts, such as by the amplification of loci from single sperm to measure recombination distances in genetic maps (40). Each successive technological innovation in DNA manipulation has ultimately been synergistic with previously existing technologies.
The concept of engineering is closely tied to rational design, a process in which scientific principles are used to design a thing and then construct it. DNA engineering technology has made it possible, and in many cases even simple and economically feasible, to create essentially any gene sequence desired. Many fundamental questions remaining on this level relate to determining what gene sequences are desired, particularly with respect to protein engineering, a very important application of gene manipulation. Although existent technology allows the synthesis of essentially any gene sequence, our current understanding of the relationships between protein structure and function does not allow the de novo design of complex gene products. Many, if not the vast majority, of genetic engineering projects currently modify a naturally occurring gene or protein for a practical or experimental purpose.
Natural evolution tends to produce proteins and other molecules that are optimized for some function in their cellular environment. Genetic variants become stabilized in a population by natural selection of beneficial traits associated with the variant. The history of DNA engineering by conventional genetics has often involved the search for naturally occurring genetic variants that convey a particularly desirable phenotype. Much effort in pharmacology and other fields has gone into identifying natural substances that can be used to serve some man-made purpose. Technology has now developed to the point that many useful or potentially useful molecules, such as plastics, are synthetic products that do not normally exist in nature. Because of this, gene and protein models for enzymes to manipulate these compounds are unlikely to be discovered as natural genetic variants. Other genetic variants useful for specialized applications may be detrimental in vivo, such as lethal autoantibodies, decreasing the probability of finding these sequence variants in natural systems.
Linking a genotype with a selectable phenotype is a key component of establishing genetic variants by natural selection. This type of linkage, which will likely be a key to developing genetic variants that do not exist naturally, can be accomplished experimentally on a molecular basis by selecting nucleic acid molecules on the basis of some characteristic determined by their sequence, such as the ability to bind a target molecule (Chapter 9), or by packaging a gene together with the product it encodes in a viral particle and selecting the particles based on some characteristic of the encoded protein (Chapters 11-12). Molecules that bind to the target of interest can then be amplified, modified, and re-selected, leading to progressively developed phenotype. This type of approach can be thought of as molecular evolution.
The development of in vitro evolution in a sense brings genetic engineering technology full circle. Random recombination of PCR products during co-amplification (41, 42) can be used to perform “sexual PCR” (43), in which “random” crossovers at points of homology between co-amplified PCR products are used as an in vitro approximation to sexual recombination. DNA engineering methods provide the means to use selections analogous to plant or livestock breeding at the molecular level, as a form of molecular husbandry in which variants are generated through mutation and recombination and those that best serve the desired function, such as affinity to a target, are selected and propagated. This is essentially genetic breeding of molecules, and is analogous to old-fashioned plant and animal breeding.
From the perspective of genetic manipulation, biological macromolecules might be viewed as micromachines subject to engineered alteration, or their individual genes can be viewed as molecular breeding stock that can be selected in vitro for specific desired characteristics. Analogously, it is important to emphasize that entire organisms can be viewed as living machines that can be engineered through the introduction of transgenes, engineered DNA molecules that are transferred into an organism. Whether present on an extrachromosomal vector or integrated into a cellular chromosome, transgenes can be used to introduce new heritable characteristics into an organism while circumventing normal biological constraints. This provides a means of genetic engineering in the broad sense of creating designed alterations of organisms.
DNA technology has now developed to the point that the ability to manipulate and synthesize gene sequences has greatly exceeded basic understanding of protein gene products. This limits application of the engineering metaphor. Molecular husbandry approaches, which view molecules as breeding stock, do not require a preconceived understanding of their functional mechanisms. We can anticipate synergy between these approaches. Semi-rational design (see Chapter 12), in which one incorporates what one knows about the molecule and lets selection do the rest, mixes the two metaphors. Molecules generated through in vitro evolution can serve as both models to be further engineered, and as experimental systems to increase our understanding of the rules relating protein sequence to protein structure. The future of genetic engineering will undoubtedly lead to many truly novel protein machines and their use as intrinsic components of biological systems.
Open Access Biology | Molecular Biology Highveld | Molecular Biology Gateway | PCR | PCR Protocols | Real-Time PCR |