O

Correct Genomic Targeting

FIGURE 4 Construction of a gene replacement mouse model. A basic gene replacement construct consists of two genomic homology arms (~1—5kb) derived from sequences flanking the exon to be deleted (1), an intervening marker (e.g., neomycin resistance, neor) for positive selection in embryonic stem (ES) cell culture, and a distal marker (e.g., thymidine kinase, TK) for negative selection in ES cells. The construct is transfected into ES cells, and the cells are grown in media containing neomycin (positive selection) and gancyclovir (negative selection). Cells with random insertions (3) retain both the neor marker and the TK marker and are killed by the conversion of gancylovir to a toxic product by TK. Correct homologous recombinants (4) retain the neor marker but lose the TK marker and thus survive the selection. Clones with correctly targeted genes are injected into donor blastocysts (5), which are then implanted into surrogates to give birth to mice chimeric for the mutation. The chimeric mice are bred to produce mice carrying the mutation in their germline.

FIGURE 4 Construction of a gene replacement mouse model. A basic gene replacement construct consists of two genomic homology arms (~1—5kb) derived from sequences flanking the exon to be deleted (1), an intervening marker (e.g., neomycin resistance, neor) for positive selection in embryonic stem (ES) cell culture, and a distal marker (e.g., thymidine kinase, TK) for negative selection in ES cells. The construct is transfected into ES cells, and the cells are grown in media containing neomycin (positive selection) and gancyclovir (negative selection). Cells with random insertions (3) retain both the neor marker and the TK marker and are killed by the conversion of gancylovir to a toxic product by TK. Correct homologous recombinants (4) retain the neor marker but lose the TK marker and thus survive the selection. Clones with correctly targeted genes are injected into donor blastocysts (5), which are then implanted into surrogates to give birth to mice chimeric for the mutation. The chimeric mice are bred to produce mice carrying the mutation in their germline.

enzymatic active site, an ion channel pore-forming region, the DNA-binding region of a transcription factor, among others. In some cases, the promoter can be targeted to disrupt gene function; however, because cryptic promoters may be present that can take over and drive expression even after the core promoter is deleted, protein coding regions of genes are generally considered better targets. Next a gene replacement plasmid is constructed that contains four basic elements: a 5' homology arm, a 3' homology arm, a positive selection cassette, and a negative selection cassette. The homology arms, which are the critical component regulating homologous recombination, flank the exon(s) targeted for deletion. The size of these arms may vary, but both are usually larger than 1 kilobase (kb). A typical construct might contain one arm that is 1kb long and the other of 3 to 5kb long. Larger homology arms may not improve targeting efficiency appreciably, but it may be helpful to select arms that do not contain repetitive elements such as L1, B1, or B2 retrotransposons. Homology arms may contain exons or other functional elements in the gene that are not targeted for deletion. It is arguable whether or not the DNA cloned into the targeting construct should be isogenic with the strain of embryonic stem cell line, but this may improve efficiency.

The selection cassettes are meant to function in cultured embryonic stem cells and usually consist of a strong, ubiquitous promoter, such as the phosphoglycerate kinase (PGK) or herpes simplex virus (HSV) promoter, driving expression of the selectable marker. The purpose of both markers is to allow identification of embryonic stem cell clones that have undergone successful homologous recombination. The positive selection marker (usually the neomycin resistance gene, neor) allows for survival of embryonic stem cells that have integrated the targeting construct (by either homologous or nonhomologous recombination) when the cells are grown in media containing neomycin, or G418. The positive selection marker is cloned between the two homology arms. The negative selection marker enriches for embryonic stem cells that have incorporated the targeting construct by homologous recombination, and it eliminates those that integrated the targeting construct by random, nonhomologous insertion. A common negative selection marker is thymidine kinase (TK), which converts gancyclovir added to the embryonic stem cell culture medium into a toxic product. The negative selection marker is cloned at one end of the targeting vector. On random, nonhomologous insertion (like a typical transgene), the entire vector is inserted, including both the neor and TK selectable markers. In cells that have undergone homologous recombination, only the positive selection marker (neor) is inserted; the negative selection marker (TK) lies outside the region of homology and does not integrate into the chromosome. Many investigators, however, do not use negative selection cassettes in their targeting construct, arguing that it does not significantly increase the percent of embryonic stem cell clones that are accurately targeted.

The gene replacement plasmid is transfected into embryonic stem cells by electroporation or lipofection, and (in our example) the cells are selected in media containing G418 (positive selection) and gancyclovir (negative selection). Genetically altered mice can exhibit variable phenotypes as a result of differences in genetic background, and this feature is partly controlled by the choice of embryonic stem cell strain. Historically most mouse strains proved refractory to the derivation of embryonic stem cell lines (Gardner and Brook, 1997), and investigators were limited to using only one or two strains, such as 129/SvJ or C57BL/6J. However, recent advances in cell culture technology are overcoming the biological barriers and permitting the efficient derivation of embryonic stem cell lines from additional mouse strains (Schoonjans et al., 2003). Following selection, the genomic DNA of surviving targeted ES cell clones is tested by PCR or Southern blot analysis to verify that the correct homologous recombination event has occurred. This assay is a critical step; sometimes only a small fraction of clones (<5%) will be accurately targeted even though those cells survived the selection process. Some investigators perform a kary-otype analysis of targeted clones to identify chromosomal rearrangements or polyploidy-aneuploidy events that may have occurred in culture.

Correctly targeted embryonic stem cells are microin-jected into normal donor mouse blastocysts, where they mix with the population of normal embryonic stem cells that constitute the inner cell mass of the early embryo. The injected blastocysts are implanted into surrogate females, where they implant in the uterus and develop until birth about 18 to 20 days later. These mice will be chimeric; that is, their bodies will be composed of a mixture of cells derived from the targeted embryonic stem cells and from the normal recipient blastocyst. Usually the targeted embryonic stem cells and the recipient blastocysts are derived from mouse strains with different coat colors so that when the chimeric mice are born, they are visibly chimeric (e.g., black and white patched coats). These chimeric mice may or may not show a phenotype, depending on the function of the targeted gene, whether or not it acts in a cell-autonomous fashion, the overall percentage of cells in the chimeric animal that are targeted, and which tissues these cells constitute. It is also important to remember that the targeted cells in a chimeric animal are heterozygous for the mutation. However, methods do exist to knockout both alleles of a gene during the embryonic stem cell culture phase, utilizing two different positive selection markers and a sequential transfection-selection approach so that targeted cells in the chimeric mice are homozygous for the mutation. The chimeric mice are then bred with wild-type mice. For those chimeras where the targeted embryonic stem cells had populated the germ tissue, 50% of the offspring will be heterozygous for the mutation in every cell. These heterozygotes are then bred to produce wild-type offspring that are heterozygous for the targeted alteration or homozygous for the alteration (with a 1:2:1 ratio). Alterations from this ratio might be observed if the targeted gene affects prenatal viability.

Variations of the Basic Gene Replacement Strategy

Numerous variations of the basic gene replacement strategy have been developed to add efficiency, flexibility, and power to the approach (Figure 5). Many of these rely on use of the Cre-loxP or flipase-frt site-specific recombinase (SSR) systems. SSRs are enzymes that recognize specific DNA sequence motifs and recombine the DNA between them. If two recombinase recognition sites are placed on the same DNA strand in the same orientation, the recombinase will delete the sequences between them. If the sites are oriented in the opposite directions, the recombinase will invert the sequences between them. The distance between the two recognition sites may be very short (<100 base pairs [bp]) or very large (>100kb). If the sites are placed on different DNA fragments, even on different chromosomes, the recom-binase will mediate a translocation event. Cre recombinase is a native component of the P1 bacteriophage, while Flp recombinase (flipase) was isolated from Saccharomyces cerevisiae (yeast). The Cre and Flp recombinases recognize 34-bp target sites, called loxP and frt, respectively. Their small size allows loxP and frt sites to be placed into a gene intron, promoter, or untranslated region without noticeably disrupting gene function. One use of the SSR system is to create an excisable selection cassette (see Figure 5). The presence of a positive selection marker (e.g., neo1) is required for selection in embryonic stem cells, but is often undesirable after the knockout is established. If the selection cassette is flanked by loxP sites, it can be deleted (after a viable targeted embryonic stem cell clone is identified but before they are injected into blastocysts) by the transient expression of Cre recombinase in those cells. All that remains at the site formerly occupied by the selection marker is one copy of the 34-bp recognition site.

Another important use for an SSR system is to produce a gene knock-in. With a knockout, the goal is to obliterate gene function completely, so it is largely irrelevant whether large fragments of foreign DNA are incorporated into the locus in the homologous recombination process. However, in a gene knock-in, the goal is to replace an endogenous exon with a mutant version of the same exon and to allow the gene to continue to function so that the effect of that mutation on protein expression can be assessed. In this case removal of the positive selection cassette is absolutely critical because even if it is located in an intron, it will significantly disrupt either or both expression and splicing patterns of the gene. Positioned correctly, so as not to inter fere with intronic enhancer elements or splicing donor or acceptor sites, the residual 34-bp recognition motif should not interfere with gene function. One benefit of producing a knock-in construct is that once a correctly targeted embryonic stem cell clone is identified, it can be used to produce both a mild knock-in and a severe knockout allele by either expressing Cre or not before injection of the cells into blastocysts.

The SSR systems can also be used to create inducible (i.e., "conditional") gene knockout models. The goal with an inducible knockout is to produce a targeted allele where a critical exon(s) is flanked by two loxP (or frt) recognition sites that do not interfere with normal gene function. This requires a much more complicated series of manipulations than necessary for a basic gene replacement, including the employment of two distinct SSR systems. First a targeted allele must be produced and the selectable marker removed (e.g., using the flipase-frt system) while the targeted exon remains intact but flanked with different SSR recognition sites (e.g., loxP sites). These embryonic stem cells are microinjected to produce a mouse with a normally functioning, but silently targeted, allele. A gene with two loxP sites in place, waiting to be recombined to delete an exon, is said to be floxed. A floxed gene can be disrupted, or "turned off," by later expressing the second recombinase (e.g., Cre) in the mouse. The power of the inducible knockout is that the Cre can be expressed in only some of the cells of the floxed mouse, or only at a specific stage of development, if desired. There are two general strategies to express Cre in a floxed mouse: either as a transgene or, less commonly, as delivered by a viral vector. In either case, the approach is limited only by the availability of well-characterized tissue-specific promoters that can be used to drive Cre expression. Alternatively Cre can be expressed under the control of an inducible promoter, such as the tetracy-cline-inducible promoter described previously, to add an additional level of spatiotemporal control over recombination of the floxed gene. One of the drawbacks of conditional knockouts is that 100% recombination of the targeted gene is rarely achieved (depending on what method or promoter is used to drive Cre expression), but the levels achieved are usually acceptable for most purposes. Additional resources on the theory and method of gene replacement, including detailed protocols, include the following: Joyner (2000), Turksen (2001), Tymms and Kola (2001), and Nagy et al. (2002).

Examples

As of this writing, 26 genes encoding ion channels, transporters, or other critical components of neurotransmission pathways have been altered by gene replacement in the mouse (see Table 1). Of these, spontaneous seizures were reported in 15 cases, a reduced threshold to chemically or

Gene Replacement (Knockout) Approaches to Modeling Epilepsy

Genomic target

Was this article helpful?

0 0

Post a comment