Recombinant DNA technology has resulted in the identification of many disease-related genes. To advance the understanding of the dis ease related to a previously unknown gene, the function of the protein encoded by that gene must be verified or identified, and the way changes in the gene’s expression influence the disease phenotype must be characterized. Analysis of the role of these genes and their encoded proteins was made possible by the development of recombinant DNA technology that allowed the production of mice that are genetically altered at the cloned locus. Mice can be produced that express an exogenous gene and thereby provide an in vivo model of its function. Linearized DNA is injected into a fertilized mouse oocyte pro nucleus and reimplanted in a pseudopregnant mouse. The resultant transgenic mice can then be analyzed for the phenotype induced by the injected transgene. Placing the gene under the control of a strong promoter that stimulates expression of the exogenous gene in all tis sues allows the assessment of the effect of widespread overexpression of the gene. Alternatively, placing the gene under the control of a regulatory sequence that can function only in certain tissues (a tissue specific promoter) elucidates the function of that gene in a particular tissue or cell type. A third approach is to study control elements of the gene by testing their capacity to drive expression of a “marker” gene that can be detected by chemical, immunologic, or functional means. For example, the promoter region of a gene of interest can be joined to the cDNA encoding green jellyfish protein and activity of the gene assessed in various tissues of the resultant transgenic mouse by fluorescence microscopy. Use of such a reporter gene demonstrates the normal distribution and timing of expression of the gene from which the promoter elements are derived. Transgenic mice contain exogenous genes that insert randomly into the genome of the recipient. Expression can thus depend as much on the location of the insertion as it does on the properties of the injected DNA.

In contrast, any defined genetic locus can be specifically altered by targeted recombination between the locus and a plasmid carrying an altered version of that gene (Fig. 1). If a plasmid contains that altered gene with enough flanking DNA identical to that of the normal gene locus, homologous recombination can occur, and the altered gene in the plasmid will replace the gene in the recipient cell. Using a mutation that inactivates the gene allows the production of a null mutation, in which the function of that gene is completely lost. To induce such a mutation, the plasmid is introduced into an embryonic stem cell, and the rare cells that undergo homologous recombination are selected. The “knockout” embryonic stem cell is then introduced into the blastocyst of a developing embryo. The resultant animals are chimeric; only a fraction of the cells in the animal contain the targeted gene. If the new gene is introduced into some of the germline cells of the chimeric mouse, then some of the offspring of that mouse will carry the mutation as a gene in all of their cells. These heterozygous mice can be further bred to produce mice homozygous for the null allele.

Fig1. GENE “KNOCKOUT” BY HOMOLOGOUS RECOMBI NATION. A plasmid containing genomic DNA homologous to the gene of interest is engineered to contain a selectable marker positioned so as to disrupt expression of the native gene. The DNA is introduced into embryonic stem cells, and cells resistant to the selectable marker are isolated and injected into a mouse blastocyst, which is then implanted into a mouse. Offspring mice that contain the knockout construct in their germ cells are then propagated, yielding mice with heterozygous or homozygous inactivation of the gene of interest.

Knockout mice reveal the function of the targeted gene by the phenotype induced by its absence. Methods for “knocking in” a gene have been developed to allow one to assess the functional consequences of replacing the function of the knocked-out gene with a modified version of that gene or an alternative gene with a related function. Genetically altered mice have been essential for discerning the biologic and pathologic roles of large numbers of genes implicated in the pathogenesis of human disease. These methods were originally developed in mice, but they have been extended to many animal species. The methods are now refined enough to generate recombinant organisms in which multiple endogenous genes are replaced by human genes, generating model organisms “humanized” for certain key functions (e.g., hemoglobin synthesis in mouse erythroid cells, certain immune functions); these humanized models are proving useful for preclinical testing of novel therapeutics.