The goal of modern molecular cell biology is nothing short of understanding the biochemical, cellular, and organismal functions of all the proteins encoded in the genome. In the preceding sections, we have discussed the isolation and analysis of mutants, the genetic mapping of mutations, and finally the isolation and cloning of mutation-defined genes. This approach can provide valuable information about the molecular mechanisms underlying the cellular processes affected by the original mutations and the in vivo functions of the normal proteins encoded by the affected genes.Many genes have been identified based on the biochemical properties of their encoded protein, the sequence similarity of the encoded protein with proteins of known function, or their interesting patterns of expression in development. In the absence of mutant forms of such genes, their in vivo functions may be unclear. By mutating a specific gene in vitro and then replacing the normal copy in the genome with a mutant form, scientists can assess its in vivo function. This technique, referred to as gene-targeted knockout, or simply "knockout," is in essence the reverse of the approach described in the previous sections. The process of isolating normal genes to be mutated will be greatly simplified as sequencing of the genomes of several model organisms and of the human genome progresses. Whether starting from a normal protein or sequenced genome, this approach can be summarized as follows:



Other techniques permit the introduction of foreign genes or altered forms of an endogenous gene into an organism. For the most part, these techniques do not result in replacement of the endogenous gene, but rather in the integration of additional copies of it. Such introduced genes are called transgenes; the organisms carrying them are referred to as transgenics. Transgenes can be used to study organismal function and development in a variety of different ways. For instance, genes that are normally expressed at specific times and places during development can be genetically engineered in vitro to be expressed in different tissues at different times and then reintroduced into the animal to assess the cellular and organismal consequences. For example, the Antennapedia (Antp) gene in Drosophila normally controls leg development, but misexpression of this gene in the developing antenna transforms it into a leg.
The production of both gene-targeted knockout and transgenic animals makes use of techniques for mutagenizing cloned genes in vitro and then transferring them into eukaryotic cells. We briefly describe these procedures first, then discuss the production and uses of knockout and transgenic organisms.
Specific Sites in Cloned Genes Can Be Altered in Vitro
Specific sequences in cloned genes can be altered in vitro and then introduced into experimental organisms. This approach has been exploited primarily to study two questions. First, what is the relationship between the structure of a particular protein and its biological function? And second, what are the specific DNA sequences required to determine the expression pattern of a particular gene?
A variety of enzymatic and chemical methods are available for producing site-specific mutations in vitro. In recent years, however, the most common methods use specific oligonucleotides as mutagens. Because oligonucleotides of any desired sequence can be chemically synthesized (see Figure 7-9), oligonucleotide-based mutagenesis can generate precisely designed deletions, insertions, and point mutations in a DNA sequence. Figure 8-29 illustrates the use of this strategy to produce a deletion.
DNA Is Transferred into Eukaryotic Cells in Various Ways
Production of both knockout and transgenic organisms requires the transfer of DNA into eukaryotic cells. Many types of cells can take up DNA from the medium. Yeast cells, for instance, can be treated with enzymes to remove their thick outer walls; the resulting spheroplasts will take up DNA added to the medium. Plant cells also can be converted to spheroplasts, which will take up DNA from the medium. Cultured mammalian cells take up DNA directly, particularly if it is first converted to a fine precipitate by treatment with calcium ions. Another popular method for introducing DNA into yeast, plant, and animal cells is called electroporation. Cells subjected to a brief electric shock of several thousand volts become transiently permeable to DNA. Presumably the shock briefly opens holes in the cell membrane allowing the DNA to enter the cells before the holes reseal. DNA also can be injected directly into the nuclei of both cultured cells and developing embryos.
Once the foreign DNA is inside the host cell, enzymes that probably function normally in DNA repair and recombination join the fragments of foreign DNA with the host cell's chromosomes. Since only a relatively small fraction of cells take up DNA, a selective technique must be available to identify the transgenic cells. In most cases the exogenous DNA includes a gene encoding a selectable marker such as drug resistance. The introduced DNA can insert into the host genome in a highly variable fashion showing no site specificity, can replace an endogenous gene by homologous recombination, or can remain as an independent extrachromosomal DNA molecule referred to as an episome.
Normal Genes Can Be Replaced with Mutant Alleles in Yeast and Mice
Gene knockout is a technique for selectively inactivating a gene by replacing it with a mutant allele in an otherwise normal organism. This technique of disrupting gene function, which has been widely used in yeast and mice, is a powerful tool for unraveling the mechanisms by which basic cellular processes occur.
Gene Knockout in Yeast
After foreign, or exogenous, yeast DNA is taken up by diploid yeast cells, recombination generally occurs between the introduced DNA and the homologous chromosomal site in the recipient cell. Because of this specific, targeted recombination of identical stretches of DNA, called homologous recombination, any gene in yeast chromosomes can be replaced with a mutant allele (Figure 8-30). The resulting heterozygous yeast cells, carrying one mutant allele and one wild-type allele, generally grow normally. To determine whether the knocked-out gene controls an obligatory function, recombinants containing the mutant allele on one chromosome are treated to induce meiosis and sporulation; each diploid cell produces four haploid spores, which are tested for viability. One of the first genes tested in this way was the one encoding actin, a prominent cytoskeletal protein in yeast and higher organisms. Haploid yeast spores without a normal actin gene cannot grow (Figure 8-31).
This technique also is useful in assessing the role of proteins identified solely on the basis of DNA sequence. For instance, the entire sequence of the S. cerevisiae genome has been determined. As described in Chapter 7, analysis of genomic sequences can identify stretches of DNA that exhibit long open reading frames or homology to genes encoding known proteins; such stretches are likely to be transcribed and translated into as yet unidentified proteins (see Figure 7-31). Gene knockouts can be used to determine whether such regions are important for specific cellular functions that are phenotypically detectable. This technique thus provides a powerful approach to identifying and studying new genes and the proteins that they encode.
This gene-knockout approach already has been used to analyze yeast chromosome III. Analysis of the DNA sequence indicated that this chromosome contains 182 open reading frames of sufficient length to encode proteins longer than 100 amino acids, which is assumed in this analysis to be the minimum length of a naturally occurring protein. The sequences of the proteins that could be encoded by 116 of these putative protein-coding regions exhibited no obvious homology to any known proteins. Gene knockout of 55 of these regions showed that 3 were required for viability; further analysis of 42 nonessential genes revealed that 14 showed a mutant phenotype and 28 did not. The large number of putative genes with no detectable mutant phenotype is quite surprising. In some cases the lack of a phenotype could indicate the existence of backup or compensatory pathways in the cell. Alternatively, the mutations may give rise to subtle defects that would require more in-depth phenotypic analysis to uncover.
Gene Knockout in Mice
Gene-targeted knockout mice are a powerful experimental system for studying development, behavior, and physiology; they also may be useful model systems for studying certain human genetic diseases. The procedure for producing gene-targeted knockout mice involves the following steps:

1. Mutant alleles are introduced by homologous recombination into embryonic stem (ES) cells.

2. ES cells containing a knockout mutation in one allele of the gene being studied are introduced into early mouse embryos. The resultant mice will be chimeras containing tissues derived from both the transplanted ES cells and the host cells. These cells can contribute to both the germ-cell and somatic-cell populations.


3. Chimeric mice are mated to assess whether the mutation is incorporated into the germ line.

4. Mice each heterozygous for the knockout mutation are mated to produce homozygous knockout mice.

The isolation and culture of embryonic stem cells, which are derived from the blastocyst, are illustrated in Figure 8-32. These cells can be grown in culture through many generations. Exogenous DNA containing a mutant allele of the gene being studied is introduced into ES cells by transfection. The introduced DNA recombines with chromosomal sequences in about 1 cell out of 100 (i.e., 1 percent recombination frequency). In some cells, the added DNA recombines with the homologous chromosomal site, but recombination at other chromosomal sites (i.e., nonhomologous recombination) occurs 103 104 times more frequently. The small fraction of cells in which homologous recombination takes place can be identified by a combination of positive and negative selection: positive selection to identify cells in which any recombination occurs and negative selection to remove cells in which recombination takes place at nonhomologous sites.
For this selection scheme to work, the DNA constructs introduced into ES cells need to include, in addition to sequences used to knock out the gene of interest, two selectable marker genes (Figure 8-33). One of these additional genes (neor) confers neomycin resistance; it permits positive selection of cells in which either homologous (specific) or nonhomologous (random) recombination has occurred. The second selective gene, the thymidine kinase gene from herpes simplex virus (tkHSV) confers sensitivity to ganciclovir, a cytotoxic nucleotide analog; this gene permits negative selection of ES cells in which nonhomologous recombination has occurred. Only ES cells that undergo homologous recombination (i.e., gene-targeted specific insertion of the DNA construct) can survive this selection scheme.
Once ES cells heterozygous for a knockout mutation in the gene of interest are obtained, they are injected into a recipient mouse blastocyst, which subsequently is transferred into a surrogate pseudopregnant mouse (Figure 8-34). If the ES cells also are homozygous for a visible marker trait (e.g., coat color), then chimeric progeny carrying the knockout mutation can be identified easily. These are then mated with mice homozygous for another allele of the marker trait to determine if the knockout mutation is incorporated into the germ line. Finally, mating mice, each heterozygous for the knockout allele, will produce progeny homozygous for the knockout mutation.
Cell-Type-Specific Gene Knockout in Mice
In most cases, investigators are interested in examining the effects of knockout mutations in a particular region of the mouse, at a specific stage in development, or both. Since most genes function in different parts of the organism and at different times, a knockout mouse may die or have defects in various tissues prior to the stage to be analyzed. To address this problem, mouse geneticists have devised a clever technique using site-specific DNA recombination sites (called loxP sites) and the enzyme, called Cre, that catalyzes recombination between them. The loxP-Cre recombination system is present in bacteriophage P1, but also promotes recombination when placed in mouse cells.
Homologous recombination strategies discussed in the previous section are used to obtain mice in which loxP sites are inserted so they flank the gene of interest or an essential exon. Since the inserted loxP sites are not within exons, they do not by themselves disrupt gene function. Transgenic mice also are prepared carrying the cre gene linked to a cell-type-specific promoter. As depicted in Figure 8-35, mating of these two types of mice will yield progeny that carry the gene of interest modified by insertion of flanking lox P sites and the cre gene controlled by a cell-type-specific promoter. In these mice, recombination between the loxP sites, which disrupts the gene of interest, will occur only in those cells in which the promoter is active and therefore producing the Cre protein necessary to induce the recombination.
One important example of this technique comes from studies on learning and memory. Earlier pharmacological and physiological studies had indicated that normal learning requires a specific neurotransmitter receptor, the NMDA class of glutamate receptors, in a specific region of the brain called the hippocampus. But mice in which the gene encoding an NMDA receptor subunit was knocked out died neonatally, precluding analysis of the receptor's role in learning. Cell-type-specific inactivation of the receptor was achieved by constructing mice carrying a Cre gene expressed in a subclass of hippocampal neurons and two different alleles of the receptor subunit gene, an allelle containing the loxP sites and a conventional knockout allele. These mice survived to adulthood and showed learning and memory defects, confirming a role for these receptors in normal learning and memory.
Use of Knockout Mice to Study Human Genetic Diseases

Gene knockout can produce model systems for studying inherited human diseases. Such model systems are powerful tools for investigating the nature of genetic diseases and the efficacy of different types of treatment, and for developing effective gene therapies to cure these often devastating diseases.
Recent studies on cystic fibrosis illustrate this use of the knockout technique. Cystic fibrosis, which afflicts about 1 in 2000 Caucasians, is caused by an autosomal recessive mutation in the CFTR gene (see Figure 8-17b). This gene was cloned by positional cloning strategies, and the biochemical function of its encoded protein studied. Using the human gene, researchers isolated the homologous mouse gene and subsequently introduced mutations in it. The gene-knockout technique was then used to produce homozygous mutant mice, which showed symptoms (i.e., a phenotype) similar to those of humans with cystic fibrosis. These knockout mice are currently being used as a model system for studying this genetic disease and developing effective therapies.
Foreign Genes Can Be Introduced into Plants and Animals
In the previous section we discussed techniques for replacing one form of a gene with another through homologous recombination. In this section we discuss methods for producing transgenic organisms, which carry cloned genes that have integrated randomly into the host genome.
Transgenic technology has numerous experimental applications and potential agricultural and therapeutic value. For instance, dominantly acting alleles of tumor-causing genes can be used to produce transgenic mice, thus providing an animal model for studying cancer. In Drosophila, transgenes often are used to determine whether a cloned segment of DNA corresponds to a gene defined by mutation. If the cloned DNA is indeed the gene in question, then introducing it as a transgene into a mutant fly will transform the mutant into a phenotypically normal individual. Transgenic plants may be commercially valuable in agriculture. Plant scientists, for example, have developed transgenic tomatoes that exhibit reduced production of ethylene, which promotes fruit ripening. The ripening process is delayed in these transgenic tomatoes, thus prolonging their shelf life. Finally, transgenic technology is a critical component in the burgeoning field of gene therapy for human genetic diseases.
Transgenic Mice
As noted in the discussion of knockout mice, specific integration of exogenous DNA into the genome of mouse cells by homologous recombination occurs at a very low frequency. In contrast, the frequency of random integration of exogenous DNA into the mouse genome at nonhomologous sites is very high. Because of this phenomenon, the production of transgenic mice is a highly efficient and straightforward process.
As outlined in Figure 8-36, foreign DNA containing a gene of interest is injected into one of the two pronuclei (the male and female haploid nuclei contributed by the parents) of a fertilized mouse egg before they fuse. The injected DNA has a good likelihood of being randomly integrated into the chromosomes of the diploid zygote. Injected eggs then are transferred to foster mothers in which normal cell growth and differentiation occurs. About 10 30 percent of the progeny will contain the foreign DNA in equal amounts (up to 100 copies per cell) in all tissues, including germ cells. Immediate breeding and backcrossing (parent-offspring mating) of the 10 20 percent of these mice that breed normally can produce pure transgenic strains homozygous for the transgene.
Numerous examples of the use of transgenic mice for studying various aspects of normal mammalian biology are presented in other chapters. They also provide a model system for studying disease processes. For example, many forms of cancer are promoted by normal cellular genes acting in a dominant fashion owing to their misregulated activity. Although transgenic mice carrying one of these genes, called myc, develop normally, tumors form at a high frequency. The observation that only a small number of cells expressing the transgene develop tumors supports a model in which additional genetic changes are necessary for tumors to form. These mice may provide an important tool for identifying those changes.
Transgenic Fruit Flies
Foreign DNA can be incorporated into the Drosophila germ-line genome by the technique of P-element transformation (Figure 8-37). This technique makes use of a segment of the P element, a highly mobile DNA element, which can transpose (jump) from an extrachromosomal element into a chromosome. (Mobile DNA elements are discussed in detail in Chapter 9.) Generally, this procedure results in incorporation of a single copy of the transgene into the Drosophila genome. In contrast, transgenic mice carry multiple copies of the transgene incorporated into their chromosomes. In both organisms, however, the chromosomal insertion site is highly variable.
Flies that develop from injected embryos will carry some germ cells that have incorporated the transgene; some of their progeny will carry the transgene in all somatic and germ-line cells, giving rise to pure transgenic lines. Individuals carrying the transgene are recognized by expression of a marker gene (e.g., one affecting eye color) that is also present on the donor DNA. Although the transgenes in Drosophila and mice insert in chromosomal sites different from the position of the corresponding endogenous gene, they usually are expressed in the right tissue and at the right time during development. Examples of the importance of this technology for studying development are discussed in Chapter 14.
Transgenic Plants

In nature, plant cells often live in close association with certain bacteria, which may provide a convenient vehicle for introducing cloned DNA into plants. Agrobacterium tumefaciens, for example, attaches to the cells of dicotyledonous plants and causes the formation of plant tumors known as galls. (Plants with two leaflets from each seed are called dicotyledons, or dicots; plants with one leaflet are called monocots .) This bacterium introduces a circular DNA molecule, called the Ti (tumor-inducing) plasmid, into the plant cell in a manner similar to bacterial conjugation. The plasmid DNA then recombines with the plant DNA. Since the Ti plasmid has been isolated, new genes can be inserted into it using recombinant DNA techniques and the Ti genes causing tumors can be disrupted. The resulting recombinant plasmid can then transfer desired genes into plant cells (Figure 8-38).
An especially useful characteristic of plants for transgenic studies is the ability of cultured plant cells to give rise to mature plants. Meristematic (growing) cells from dissected plant tissue or cells within excised parts of a plant will grow in culture to form callus tissue, an undifferentiated lump of cells. Under the influence of plant growth hormones, different plant parts (roots, stems, and leaves) develop from the callus and eventually grow into whole, fertile plants. When an agrobacterium containing a recombinant Ti plasmid infects a cultured plant cell, the newly incorporated foreign gene is carried into the plant genome.
As noted above A. tumefaciens readily infects dicots (petunia, tobacco, carrot) but not monocots; reliable techniques for introducing genes into monocots are still being developed. Direct introduction of DNA by electroporation has been successful in rice plants, which are monocots, and the future looks bright for the manipulation of other commercially important monocotyledonous crop plants. Also available for gene-transfer experiments are cells of a tiny, rapidly growing member of the mustard family called Arabidopsis thaliana. This plant is well-suited to genetic analysis of a variety of developmental and physiological processes. It takes up little space, is easy to grow, and has a small genome, and genes defined by mutations can be cloned by positional cloning strategies.
SUMMARYGenes can be modified in vitro by a variety of enzymatic and chemical methods. Modified genes can be incorporated into the germ line at their original genomic location by homologous recombination, producing knockouts, or at different sites by nonhomologous recombination, producing transgenics.

*
Any cloned yeast gene or mouse gene can be mutated and used to generate gene-targeted knockouts in these organisms (see Figures 8-27 and 8-31). Use of the loxP-Cre recombination system from bacteriophage P1 permits production of mice in which a gene is knocked out in a specific tissue (see Clues about the normal function of a gene can be deduced from the observed effects of disrupting it by the knockout technique. Mouse knockouts can provide models for human genetic diseases such as cystic fibrosis.
Transgenic single-cell organisms, plants, and animals can be produced readily by several different methods. These modified organisms contain one or more copies of a cloned gene integrated into in the genome.
*
The transgenic technique can be used to introduce a normal copy of a gene into a mutant organism, thereby identifying a cloned DNA corresponding to a mutation-defined gene. It also is used to study sequences necessary for gene expression, to develop mouse models of dominant forms of human diseases, to modify plants, and to investigate the relationship between the structure of a protein encoded by a gene and its function