Online Issues
<< All Back-issues
<< This Issue's Table of Contents
ILAR Journal V38(3) 1997
Animal Models of Aging Research
Arlan Richardson, Ahmad R. Heydari, William W. Morgan, James F. Nelson, Z. David Sharp, and Christi A. Walter
| Arlan Richardson, Ph.D., is Director of the Aging Research and Education Center and Professor of Physiology at the University of Texas Health Science Center at San Antonio (UTHSCSA) and Research Career Scientist at the South Texas Veterans Health Care System, San Antonio. Ahmad R. Heydari, Ph.D., is Assistant Professor in the Department of Nutrition and Food Science at Wayne State University, Detroit, Michigan. William W. Morgan, Ph.D., is Professor in the Department of Cellular and Structural Biology at UTHSCSA. James F. Nelson, Ph.D., is Associate Professor in the Department of Physiology at UTHSCSA. Z. Dave Sharp, Ph.D., is Associate Professor and Deputy Chair in the Department of Molecular Medicine and Deputy Director of the Institute of Biotechnology at UTHSCSA. Christi A. Walter, Ph.D., is Director of the Transgenic Animal Facility and Associate Professor in the Department of Cellular and Structural Biology at UTHSCSA. |
INTRODUCTION
One of the major difficulties in aging research (especially aging research using higher organisms) has been the ability to test experimentally whether a change in a specific gene or physiological process is involved in aging. For the most part, investigators have been forced to conduct correlative experiments with animals, for example, correlating changes in a process or a macromolecular entity to changes with aging. Although correlative experiments are important because they define what changes occur in an organism with age and they provide investigators with information that can be used to support or refute various theories of aging, these types of experiments do not allow one to test directly the role of a specific gene in aging or a theory of aging. With the advent of recombinant DNA technology and the ability to genetically engineer mice, investigators now have an experimental system whereby a specific gene or process can be altered in rodents and the effect of this alteration on aging can be studied. Using this unique system will make it possible to probe more directly the biological mechanism(s) underlying the aging process. In this article, we discuss the current experimental methods used in producing transgenic mice and the potential problems that investigators should consider when developing transgenic mouse models to study aging.
METHODS FOR PRODUCING TRANSGENIC MICE
Transgenic animals are defined as animals that either carry a fragment of foreign DNA stably integrated into the genome of the organism or have a portion of the genome deleted or mutated. In general, these animals are produced so that the genetic alterations are stably transmitted to the progeny through the germ line. The first report describing the production of transgenic mice was published in 1980 (Turrens and Boveris 1996). Subsequently, transgenic mice have been used extensively to study a variety of biomedical questions (for review see Cruse and Lewis Jr. 1994; Ho 1994; Knapp and Kopchick 1994; Koretsky 1992; Montoliu 1994; Norwood and Gomez 1994; Rusconi 1991; StC. Sinclair 1995; Sullivan and others 1993; Wagner and others 1995). In this article, we discuss 2 experimental approaches in which transgenic mice can be used to study aging: (1) gain in function, and (2) reduction or loss of function.
Transgenic mice have been generated using 3 techniques (Hogan and others 1994). The most straightforward and widely used method is the introduction of DNA into the nucleus of a cell. This method involves the microinjection of exogenous DNA into the pronuclei of fertilized eggs as shown in Figure 1. The injected eggs are then transferred to the oviduct of a pseudopregnant female and allowed to develop to term. The integration of the exogenous DNA into the genome of the fertilized egg is believed to occur through the normal process of chromosomal breakage and repair, occurring randomly in the genome. The exogenous DNA is often integrated as multiple copies into 1 site in a head-to-tail array; however, multiple integrations also can occur. In this method of producing transgenic mice, it is important that the random integration occurs in the 1-cell stage so that the exogenous DNA is integrated into the genome of all cells in the organism, especially the germ cells. When integration occurs at later stages, mosaic animals are produced in which the exogenous DNA is confined to specific developmental compartments or cell lineages.
Transgenic mice also have been produced by the infection of early embryos with recombinant retroviruses carrying an exogenous gene. This method has an advantage in that only 1 copy of the provirus is found at the chromosomal integration site; however, the technique has several limitations. For example, the retroviruses may not uniformly infect all cells in early embryos, therefore the frequency of germ line transmission is relatively low. In addition, the expression of genes introduced in retroviral vectors is often low, and only a relatively small fragment of the DNA (smaller than 10 kb) can be cloned into the retroviral vector.
The third method for producing transgenic mice takes advantage of the pluripotential capacity of embryo-derived stem (ES1) cells. In this method, the ES cells in culture are transformed using conventional gene transfer techniques, and the transformed cells are reintroduced in early blastocysts, which are transferred to appropriately timed pseudopregnant females. The transformed ES cells contribute to the development of a chimeric offspring; one can identify chimeric mice in which the ES cells have contributed to the germ line. This method allows an investigator to preselect the desired genotype or phenotype in the cultured ES cells. Although investigators have used ES cells to produce transgenic mice that express specific genes, this method is used primarily to target a transgene construct to a predetermined chromosomal locus through homologous recombination, thus producing transgenic mice that carry a mutation in a specific gene. These mutations can inactivate the gene or alter the function of the corresponding protein. Thus, the ES cell method has several advantages over the other techniques for producing transgenic mice, although the culturing and maintenance of fully totipotent ES cells are technically demanding.
USE OF TRANSGENIC ANIMALS SHOWING GAIN IN FUNCTION IN AGING RESEARCH
The majority of studies with transgenic mice have employed models in which the animals have been engineered genetically to overexpress a specific gene product in cells that either normally express or do not express the gene. Such transgenic mice have been produced almost exclusively by the microinjection of a specific transgene into the pronuclei of fertilized eggs as shown in Figure 1. Below we discuss several factors that should be considered when producing transgenic mice for aging studies by this technique, including the transgene construct, the chromosomal site of integration, the strain of mice, and the use of transgenic rats.
Transgene Construct
Transgenic mice have been produced using DNA fragments containing an entire gene, that is, both the exons and introns of the gene of interest as well as the 5'-flanking enhancer/ promoter sequences that regulate the transcription of the gene. However, investigators more commonly use chimeric transgenes that are made up of several components as shown in Figure 2. Chimeric transgenes generally consist of the full-length cDNA for a specific gene fused to an enhancer/ promoter sequence. The transgene must also contain a polyA cleavage site (which contains the polyadenylation signal, the cleavage site, and the cleavage factor binding site) on the 3'-end of the cDNA to ensure that the mRNA transcript will be processed correctly to produce a polyadenylated transcript. However, it has been observed that cDNA-containing transgenes are sometimes poorly expressed in transgenic mice (Choi and others 1991; Palmiter and others 1991), and an intron has been shown to increase transcriptional efficiency for some cDNA constructs (for example, b-globin, metallothionein, growth hormone) 10- to 100-fold (Brinster and others 1988). Producing transgenic mice with a chimeric transgene has several advantages over an intact gene. First, the lack of introns in the cDNA means that chimeric transgenes are smaller than the intact gene. This is important for large genes because of the lower efficiency of producing transgenic mice with large DNA fragments. Second, one has the option of choosing a particular enhancer/promoter to drive the expression of the cDNA, which allows one to direct the expression of a transgene to a specific tissue or tissues and to control when the transgene is expressed in the life of the animal. In addition, it is easy to identify mice containing a chimeric transgene.
Investigators studying the mechanism of aging are faced with a choice in the selection of an enhancer/promoter to drive the expression of transgenes. Because of the global nature of the aging process (aging affects essentially all cells and no 1 tissue is responsible for aging), transgenic experiments designed to test mechanisms of aging may require that the gene or process of interest be altered in most, if not all, tissues of the transgenic mice. For example, to test the role of DNA damage in aging, it would be ideal to develop a transgenic mouse that showed enhanced DNA repair in all tissues/cells. The best promoters/enhancers for these types of aging studies would thus be those that direct the ubiquitous expression of a gene. Unfortunately, most of the promoters and enhancers that have been characterized in transgenic mice direct the expression of genes to 1 or a few tissues. Information is very limited on promoters/enhancers that are capable of directing the expression of genes to a broad spectrum of tissues. One enhancer/ promoter that has been used extensively in transgenic mice and that is often considered to give broad tissue expression is the b-actin enhancer/promoter (from a chick). Although the b-actin enhancer/promoter gives high expression of transgenes in heart, muscle, lung, and brain, its expression in liver, kidney, intestine, and spleen is very low or not detectable (Balling and others 1989; Kagan and others 1994; Nilsson and Lendahl 1993; Oberly and others 1993; Sands and others 1993; Yamashita and others 1993). The 0.6-kb human cytomegalovirus (CMV1) enhancer/promoter has been used extensively in a variety of mammalian cell lines in culture to overexpress transgenes. The human CMV enhancer/promoter gives high expression of transgenes in transfected cells, and several groups also have used the human CMV enhancer/ promoter to drive the expression of reporter genes in transgenic mice (Dressler and others 1993; Furth and others 1991, 1994; Kothary and others 1991; Schmidt and others 1990). In general, the human CMV enhancer/promoter gives relatively high expression of the transgene in many tissues (for example, muscle, heart, kidney, spleen, testis, and brain); however, expression is low or undetectable in the liver. Other limitations with the CMV enhancer/promoter include the findings of Furth and others (1991), that significant variation (up to 100,000-fold) exists in the expression of the CMV enhancer/promoter transgene in tissues of transgenic mice. In addition, it is possible that the human CMV enhancer/promoter might become inactivated by methylation in tissues of transgenic mice. Although studied to a more limited extent than the human CMV enhancer/promoter, preliminary studies suggest that the C/EBP[5 (Talbot and others 1994) and the 3-hydroxy-3-methylglutaryl CoA reductase (Duhamel-Clerin and others 1994) enhancers/promoters may give nearly ubiquitous expression of transgenes. However, these initial studies need to be confirmed by other laboratories with other transgenes. Thus, only a few promoters/enhancers currently appear to be suitable for studies testing the theories or mechanisms of aging in which the global expression of a transgene is required.
Site of Chromosomal Integration
Although the mechanism whereby a transgene becomes integrated into the genome of the fertilized egg is unknown, several characteristics of this process have been defined. For example, the chromosomal integration of transgenes is random, and linear molecules integrate more efficiently into the host's genome than circular molecules. Usually all of the molecules that integrate are on the same chromosome and at the same site, with the multiple integrated molecules usually arranged in tandem.
The site of integration is important for 2 reasons. First, the expression of the transgene is affected by the neighboring regions of chromatin. For example, if the transgene is incorporated in silent, heterochromatic regions of the genome, the expression of the transgene can be low or not detectable. To circumvent this problem, investigators routinely generate several founders and select the line of transgenic mice showing the highest expression of the transgene in the tissue or tissues of interest. Preliminary research suggests that transgenes might be insulated from the general negative effects of the flanking sequences of the genome by linking transgenes to DNA sequences containing matrix-attached regions (Boulikas 1993; McKnight and others 1992; Rusconi 1991; Stief and others 1989). These sequences are believed to assist in the reorganization of the linked transgene into active chromatin, allowing the transgene to function independently from neighboring sequences. However, the use of sequences containing matrix-attached regions in the production of transgenic mice is not well-characterized, therefore such sequences are not routinely used in generating transgenic mice.
The second and most important concern in the chromosomal integration of a transgene is whether the integration occurs at a "permissive" location in the host genome. In other words, does the insertion disturb the expression of endogenous genes'? For example, the transgene could become integrated into a gene resulting in a null mutation, or the transgene could become integrated into a region of the genome that contains pertinent regulatory elements for endogenous gene expression. The problem of permissive integration is a serious experimental concern because it limits the ability of an investigator to relate the function of the transgene to a phenotype. This occurs because the expression of endogenous genes could be altered inadvertently as a result of the integration of the transgene/targeting sequence into the mouse genome.
To eliminate the problem of permissive integration and demonstrate that the overexpression of a transgene is responsible for a certain phenotype, it is necessary to study at least 2 or 3 lines of transgenic mice generated from different founders. If the phenotype is due to the site of chromosomal integration of the transgene (that is, changes in the expression of endogenous genes and not the transgene), the phenotype would not be observed in other transgenic lines because it would essentially be impossible to find a transgene randomly integrated into the same site of the genome in 2 different founder mice. However, if the phenotype is similar for all transgenic lines generated from different founders, the evidence is conclusive that the phenotype was due to the overexpression of the transgene. Nevertheless, this is a particularly important problem when investigators study the effect of a transgene on aging because the expense of maintaining colonies of several lines of transgenic mice for survival and pathology studies would be considerable.
Strains of Mice
The selection of the strain of mice to use in producing transgenic mice is especially critical in aging research, and the strains of rodents used in aging research have been discussed previously (Hazzard 1991; Hazzard and Soban 1988; Hazzard and others 1992; Masoro 1991). A variety of hybrid and inbred strains of mice have been used to produce transgenic mice by the microinjection method shown in Figure 1 (Ho 1994; Hogan and others 1994). Most often, DNA is injected into F2 hybrid zygotes generated from matings between F1 hybrid female and F1 hybrid male mice (for example, C57BL/ 6 x SJL, C57BL/6 x DBA, C57BL/6 x C3H/He, or BALB/c x C57BL/6). The F2 founder mice that are produced are then mated to the F2 hybrid mice, producing F3 mice; and the transgenic line is generated by continued mating of the hybrid offspring giving generations F4, F5, F6, and so forth.
Because of continued recombination events with every generation, genetic variation will be considerable from mouse to mouse and generation to generation. In other words, no mouse will be genetically identical to any other mouse. This genetic diversity of hybrid mouse colonies creates a major problem for aging studies because of the difficulty in maintaining a genetically stable colony of hybrid mice. For example, continuous breeding of F2, F3, F4, and subsequent generations of mice would result in genetic drift and unplanned selection (for viability, fertility, and so forth). Therefore, it is quite likely that the survival and pathological characteristics of the colony of hybrid mice will change significantly with time. This limits the usefulness of colonies of hybrid mice for aging research because it is difficult, if not impossible, to duplicate studies because an animal colony that has an identical genetic background is not available. The problem of maintaining a genetically stable colony of hybrid mice can be minimized partially by continuously breeding the transgenic mice at each generation back to the F1 parental strain of nontransgenic mice. Such a breeding strategy would maintain the heterozygosity of a colony transgenic mice. However, once an investigator inbreeds the transgenic mice to obtain homozygous mice, genetic drift and selection will subsequently occur in the colony.
The problems of genetic variation between animals and genetic drift/selection can be minimized by using inbred strains of mice to generate transgenic mice. Although inbred transgenic mice have been produced (for example, C57BL/6, FVB/ N, BALB/c, and C3H) (Hogan and others 1994), the efficiency of generating transgenic mice in inbred strains is relatively low. For example, Brinster and others (1985) reported that the efficiency of producing transgenic mice in C57BL/6 mice was one eighth that of a hybrid strain (C57BL/6 x SJL)--3.3% compared with 27.1%. We have observed a similar difference in the efficiency of producing transgenic mice in C57BL/6 and C3HeB/Fe inbred mice compared with the F1 hybrids BALB/c x C57BL/6 and C57BL/6 x DBA2. Thus, the initial expense and time required to obtain transgenic mice with inbred lines is much greater than for hybrid mice. Nevertheless, we have found that inbred strains of mice are more likely than hybrid mice to express the transgene when the appropriate elements are present in the enhancer/promoter. Most importantly, one does not have the problem of maintaining a genetically stable colony of mice using transgenic, inbred mice. Several inbred strains of mice that are widely used in aging research can be used to produce transgenic mice (for example, C57BL/6, DBA/2, CBA/Ca, and BALB/c). Considerable information exists on the survival and pathological lesions that occur with age in these 4 inbred strains of mice (Blackwell and others 1996; Hazzard 1991; Hazzard and Soban 1988; Hazzard and others 1992; Masoro 1991). In addition, many inbred strains of mice (especially C57BL/6) are widely used in aging research, and considerable information exists on the survival and pathological lesions that occur with age in these inbred strains (Blackwell and others 1996; Hazzard 1991; Hazzard and Soban 1988; Hazzard and others 1992; Masoro 1991). In contrast, no survival/pathological data exist for colonies of hybrid mice maintained through the continued mating of the hybrids, and such colonies of hybrid mice have not been used in aging research. Because of the problems of genetic variability, genetic drift/selection, and the lack of survival/pathology data for hybrid colonies of mice, it is well worth the additional expense and effort required to generate transgenic mice in inbred strains (for example, C57BL/6) for aging research.
Transgenic Rats
At the time of this writing, almost all research using transgenic animals has employed mice because the procedures for producing transgenic mice have been well-characterized over the past decade (Cruse and Lewis Jr. 1994; Koretsky 1992; Sullivan and others 1993). Nevertheless, a few studies have produced and used transgenic rats (Dycaico and others 1994; Morimura and others 1993; Peters and others 1993). Although the genome of the mouse is much better characterized than that of the rat and the cost of producing and maintaining transgenic mice is less than for rats, there are research questions in which transgenic rats would be preferable to transgenic mice. For example, rats are widely used in physiological studies simply because of their larger size. Certain surgical techniques that are used in studying organ function are very difficult or impossible to perform in mice; or, as is the case with studies of the prostate, the size of an organ in the mouse makes it very difficult to study. In addition, a rat model might be superior to a mouse model for studies on the toxicological or carcinogenic properties of compounds because the sensitivity of rats and mice to a carcinogen or toxin might be quite different. For example, mice are relatively resistant to the action of alkylating agents: less than 10% of mice from the most sensitive mouse strain develop brain tumors after treatment with an alkylating agent (Denlinger and others 1974). In contrast, more than 90% of rats develop brain tumors when treated with an alkylating agent (Swenberg and others 1972). Therefore, if one were studying the effect of overexpressing a DNA repair gene on alkylation-induced brain tumors, a transgenic rat model would be superior to a mouse model. In addition, rats have been widely used in aging research, and a very large data base exists on the effect of aging in certain rat strains (for example, F344 rats) (Hazzard 1991; Hazzard and Soban 1988; Hazzard and others 1992; Masoro 1991). In fact, the effect of age on physiological and hormonal functions has been studied more extensively in rats than in mice. Thus for certain questions in aging, a transgenic rat model might be more appropriate than a transgenic mouse model, and investigators should seriously consider the option of using transgenic rats in these cases.
USE OF TRANSGENIC MICE SHOWING REDUCED OR LOSS IN FUNCTION IN AGING RESEARCH
Since the first report of germline transmission of a targeted mutation appeared in 1989 (Thompson and others 1989), hundreds of transgenic mice have been produced with mutations in specific genes. Using the ES cell system described above, it is feasible to target a desired mutation to any gene in the mouse genome (Hogan and others 1994). To perform gene-targeting experiments, a piece of mutated mouse genomic sequence is transfected into the pluripotent ES cells, and this mutated genomic sequence is integrated into the genome of ES cells through homologous recombination as shown in Figure 3. Technically, various types of mutations can be generated using this system. Examples are point mutations, in which a slightly altered protein product is produced, or, as is most often the case, a null (knockout) mutation (Figure 3), in which a portion of the gene has been deleted and the expression of the gene is therefore abrogated.
One can identify and propagate a pure population of recombinant ES cells containing the mutated sequence. These ES cells are then injected into blastocysts, which are transferred into the uterus of a pseudopregnant mother. Once the ES cells are returned to the environment of the early embryo, they can contribute to all cell types (including the germ line) of the resulting chimeric mouse. Thus, any genetic alteration that can be introduced into ES cells in culture can be transferred to the intact animal. By breeding mice heterozygous for the mutation, one can obtain a homozygous mouse carrying the mutated allele. However, it is not unusual to find that transgenic mice homozygous for a null mutation are not viable. For example, the fetuses die during embryonic development, or the mice survive for only a short time after birth. Therefore, lines of mice containing a null mutation are often maintained as heterozygous transgenic mice.
Transgenic Mice Homozygous or Heterozygous for Null Mutations
Transgenic mice homozygous for a null mutation have been used extensively in research to determine how the total elimination of a specific gene product alters an organism. However, in many cases transgenic mice homozygous for a null mutation are not useful in aging research because these animals have a limited survival or major pathological problems. In contrast, transgenic mice heterozygous for a mutated gene generally have been underutilized, and one often sees the comment that the heterozygous transgenic mice are "normal,'' which in reality means that the mice live to adulthood (6 to 8 months of age) and are fertile, producing viable offspring. We currently know of no case in which the survival and pathological lesions associated with age have been characterized in a heterozygous transgenic mouse, therefore it is premature in most cases to classify most heterozygous transgenic mice as normal.
Transgenic mice heterozygous for a mutated gene represent a potentially valuable, untapped resource for the aging community. In general, the expression of the mutated gene in tissues of the heterozygous mice is approximately one half that found in the tissues of wild-type mice. However, one must determine that expression of the mutated gene is reduced in tissues of transgenic mice because it is possible that cells/tissues compensate for the reduced copy number by up-regulating the expression of the nonmutated allele. Transgenic mice heterozygous for a null mutation give investigators in aging research the opportunity to study how the reduced expression of a specific gene affects survival and pathology. For example, transgenic mice with a null mutation in the Mn-superoxide dismutase (Mn-SOD1) gene have been produced recently by 2 laboratories (Lebovitz and others 1996; Li and others 1995). Although the homozygous mice die before 3 weeks of age, the heterozygous mice are viable even though tissues show a 50% reduction in Mn-SOD activity.
Because Mn-SOD plays a critical role in protecting mitochondria from reactive oxygen species, the Mn-SOD heterozygous transgenic mouse would be an interesting model for testing how the accumulation of oxidative damage in the mitochondria affects aging. One might argue that a 50% decrease in the activity of an enzyme or a biochemical process would not be a large enough change to affect aging. However, most enzymes that decrease with age show less than a 50% decrease (Finch 1972; Richardson and Birchenall-Sparks 1983; Richardson and Semsei 1987; Richardson and others 1985). In addition, the changes in gene expression that occur with dietary restriction are generally in the 30 to 50% range (Pahlavani and others 1994; Van Remmen and others 1995); dietary restriction extends the survival of rodents by retarding aging (Masoro 1984, 1985). Therefore, the changes in gene expression that occur with age or are altered by dietary restriction are in the same range as the changes in expression found in transgenic mice heterozygous for a null mutation.
Neighborhood Effects
One of the concerns in studying null mutants that are produced through the deletion of part or all of the target gene is the effect of such a disruption on other genes located near the targeted gene, that is, neighborhood effects. This concern is particularly relevant to aging research because it becomes difficult to associate changes in the function of a specific gene to a phenotype. Although this is potentially an important problem, there is little information on either the frequency or severity of possible neighborhood phenotypic effects in knockout mice. In other words, does the deletion of a major portion of a gene alter the expression of other genes? If such neighborhood phenotypic effects occur, the ability of investigators to assess cause-effect relationships with null mutants produced through gene disruption would be seriously complicated.
Recent studies show that neighborhood effects do occur in the targeted inactivation of the myogenic basic-helix-loop-helix gene MRF4 (Olson and others 1996). Three laboratories independently inactivated the MRF4 gene using different strategies by which different portions of the MRF4 gene were deleted. However, all 3 strategies resulted in a null mutation in the MRF4 gene, and no protein product was produced in these mice. The phenotypes of these 3 different null mutants ranged from complete viability to complete lethality. The evidence suggests that the different phenotypes arose because of the effect of the deletion on the expression of an adjacent gene, the Myf5 gene. In other words, it appears that regulatory elements for the Myf5 gene are present in some portions of the MRF4 gene, and when these regions of the genome were deleted, the expression of the Myf5 gene was altered and a lethal phenotype occurred. Thus, the dispersal of pertinent regulatory elements over very large stretches of DNA complicates the interpretation of phenotypic data gathered from strategies in which major portions of a gene are deleted.
The studies with the MRF4 gene knockouts highlight the importance of knowing as much as possible about the entire gene to be targeted, including its various regulatory elements and the identity and regulation of adjacent genes. To avoid or minimize the problem of deleting sequences that may have regulatory effects on adjacent genes, one can introduce an effectively positioned stop codon into a coding sequence. When this strategy is coupled with the removal of the selection cassette by site-specific recombinases, one can obtain the cleanest introduction of a small and specific mutation that should have minimal effect on the expression of adjacent genes. Although such a strategy requires greater effort in establishing and identifying the desired ES cell line, it appears to be well worth the effort.
Strains of Mice
The ability to maintain ES cell lines so that they retain a normal karyotype is a critical factor in the ability of ES cells to form functional germ cells with high efficiency, and this feature limits the strains of mice that can be used as a source of ES cells. Currently, most ES cell lines are derived from the 129 strain of mice (Hogan and others 1994), which is homozygous wild-type (A/A) at the agouti locus. These ES cells are injected into blastocysts from an albino or homozygous nonagouti (a/a) strain such as C57BL/6, a procedure that allows one to recognize chimeric mice by coat color. If the chimeras are mated to 129 mice, the mutation will immediately be on an inbred background. Most often, however, the chimeras are mated to another inbred strain such as C57BL/6, which results in the mutation on a hybrid background. Several substrains of 129 mice currently exist, and they appear to vary both genetically and phenotypically. Unfortunately, investigators in many cases are not aware of the substrain of 129 mice used as the source of ES cells. Although it has been suggested that ES cells from C57BL/6 mice can also be used to produce transgenic mice (Hogan and others 1994), such cells have been used to a very limited extent, and it appears that problems with genetic instability may limit their usefulness.
One major problem in using null mutants in aging research is that essentially nothing is known about the 129 strain and substrains of mice with respect to aging. Except for an early study by Smith and others (1973) in which the survival and tumor incidence of the 129 strain of mice were measured, there is only very limited information on the pathology of the 129 substrains (Coulson and Wilson 1989; Schlager 1966; Staats 1972, 1981; Stevens 1973), and to our knowledge there are no aging studies with the 129 strain or substrains of mice. To circumvent the problem of using the poorly characterized 129 substrains of mice for aging studies, one should consider backcrossing the null mutant lines to an inbred line that has been well-characterized with respect to aging (for example, C57BL/6). After backcrossing for 20 generations, the resulting mice would be congenic and homozygous C57BL/6 at virtually all loci except the mutated locus and the surrounding sequences, which would be derived from the ES cells (that is, the 129 strain). Although this currently appears to be the best approach for aging research, it is still not ideal because the genes surrounding the mutation that are derived from the 129 strain could contribute to the phenotype, thereby complicating the interpretation of survival and pathological data. In other words, it would be possible for the changes in survival and pathology to arise not from the mutated gene but instead from an adjacent gene from the 129 genome.
IMPORTANCE OF TEMPORAL EXPRESSION OF TRANSGENES IN AGING RESEARCH
Current transgenic models are limited because the transgenes either are continuously and aberrantly active or are suppressed throughout the life span of the animal. This problem is particularly acute during prenatal development, when the embryo may simply die or alternatively compensate, sometimes remarkably, for the aberrant expression of the manipulated gene. An example of this type of compensation is the apparently normal development of skeletal muscle in transgenic mice with a knockout of the MyoD gene, a key gene in muscle determination (Rudnicki and others 1992). This unanticipated response appears to arise from a compensatory increase in Myf-5 expression during development (Rudnicki and others 1993). Therefore, it is not always easy to define the normal physiological contribution of specific genes to complex lifelong processes (such as aging) by the application of conventional transgenic mouse approaches. The ability to express a transgene temporally after the mice have reached maturity (for example, 6 months of age) would be advantageous in aging research because it would allow so-called normal levels of gene expression during prenatal development. In addition, the ability to temporally express a transgene would allow one to define the stages of life when the expression of the gene would have the greatest effect on aging.
Over the past 2 decades, several inducible promoters have been used to temporally express transgenes. These are naturally occurring promoters that are activated by a variety of agents, such as heavy metals (Mayo and others 1982), heat shock (Wurm and others 1986), and glucocorticoids (Heynes and others 1981; Lee and others 1991). In general, these promoters have proven to be unsatisfactory for 2 major reasons. First, an unacceptable level of background expression of the transgene is observed in the absence of the inducer. Second, heavy metals, heat shock, and glucocorticoids induce a variety of endogenous genes, which could complicate the interpretation of data using these inducing agents.
Binary Expression Systems
A new and particularly promising approach for regulating the activity of specific genes in an intact animal model involves using genetically engineered binary expression systems. A generalized example of such a system is shown in Figure 4. The special features of the binary expression systems that make them ideal for the temporal expression of transgenes in mammals are (1) a transactivator-dependent promoter that is unique and will not respond to transcription factors found in mammalian cells; (2) a genetically engineered transactivator protein that will specifically bind to the unique, transactivator-dependent promoter and activate transcription in mammalian cells; and (3) a repressor or activator that specifically inhibits or induces the binding of the transactivator protein to the transactivator-dependent promoter but has no, or little, effect on the expression of endogenous mammalian genes.
The way in which the 2 transgenic components of a binary system interact to control the expression of a gene is shown in Figure 4. The first component is a chimeric transactivator gene, which is under the control of a promoter that will express the transactivator in the cells and tissues of interest. The chimeric transactivator gene codes for a protein that contains both DNA binding and transcriptional activating domains. The DNA binding domain will allow the transactivator protein to bind to a specific, unique DNA sequence (that is, it is not found in the promoters of endogenous mammalian genes), and the transcriptional activating domain of the protein will interact with the transcriptional initiation complex in mammalian cells to activate transcription. The second component of the binary system is the gene of interest placed under the control of a transactivator dependent promoter (that is, a promoter containing the unique DNA sequence recognized by the DNA binding domain of the transactivator protein). Two lines of transgenic mice that contain each of the constructs are generated, and transgenic mice containing both constructs in their genome are produced by mating the 2 lines of transgenic mice.
Two types of binary systems that regulate the temporal expression of transgenes have been developed. In the first system, the transactivator protein spontaneously binds to the transactivator-dependent promoter and induces the expression of the transgene. In other words, the expression of the transgene is turned on under normal situations in the tissues of the transgenic mice that express the chimeric transactivator. However, in the presence of a repressor, the binding of the transactivator protein to the transactivator-dependent promoter is inhibited, and the expression of the transgene is repressed. In the second type of binary system, the transactivator protein will bind only to the transactivator-dependent promoter in the presence of an activator, that is, the expression of the transgene is normally turned off and is induced by the addition of an activator. Thus, the binary systems allow one to time the overexpression of a specific gene in transgenic mice. In the future, it may be possible to time the knockout of a gene in transgenic mice by using the binary systems in conjunction with the cre/lox or flipase systems.
Tetracycline Binary System
Gossen and Bujard (1992) developed a binary expression system that is based on the phenomenon of tetracycline resistance in Escherichia coli, and the specific components of this system are shown in Figure 5A. The tetracycline-repressible binary system employs a promoter containing a heptad of the E. coli tetracycline operator sequence. Transgenes fused to this promoter would essentially be silent in mammalian cells because of the lack of a transcription factor (for example, the tetracycline-repressor protein) that can bind the promoter and activate the transcription of the transgene. The genetically engineered chimeric transactivator used in the tetracycline binary system is also shown in Figure 5A and consists of the E. coli tetracycline-repressor gene fused to the DNA sequence coding for the carboxy terminal domain of the herpes simplex V16 protein. The tetracycline-repressor domain of the chimeric transactivator will bind to the heptad sequences of the tetracycline operator, and the carboxy terminal domain of the herpes simplex V16 gene will interact with the transcriptional machinery of mammalian cells and initiate the transcription of the transgene. Tetracycline acts as a repressor in the tetracycline binary system. It binds with high affinity to the tetracycline-repressor domain of the transactivator protein, and this binding induces allosteric changes, which prevents the transactivator from binding to the tetracycline operator. Because tetracycline-repressible gene expression is found in bacteria and not in mammalian cells, the administration of tetracycline to mammalian cells has little, if any, detectable pleiotrophic effect on endogenous gene expression.
In 1994, Furth and others demonstrated that the tetracycline binary system could be used in transgenic mice. In that study, the chimeric transactivator was placed under the control of the human CMV promoter, and temporal expression was measured for a reporter gene fused to the promoter containing the tetracycline operator shown in Figure 5A. When slow-release tetracycline pellets were implanted subcutaneously into mice, the expression of the reporter gene was essentially at background levels, and this suppression was effective at a dose of the antibiotic that had no demonstrable toxic effects on either fetal development or adult transgenic mice. In the absence of tetracycline, the expression of the reporter gene was 50- to 150-fold above background levels. Three other groups have used the tetracycline binary system to temporally express transgenes in transgenic mice. Passerman and Fishman (1994) generated transgenic mice in which the chimeric transactivator was fused to the rat a-myosin heavy chain promoter and the tetracycline promoter fused to the Id1 gene. The heart-specific expression of the Id1 gene was completely repressed by tetracycline administration. In a second study, the insulin promoter was used to drive the expression of the chimeric transactivator specifically in pancreatic b-cells of transgenic mice (Efrat and others 1995). When these mice were bred to founders carrying a transgene in which the tetracycline promoter was linked to the SV40 large T-antigen, b-cell-specific tumors developed in the transgenic offspring. Cell lines derived from these tumors showed growth arrest in the presence of tetracycline but proliferated normally upon its withdrawal. More recently, Ewald and others (1996) used the tetracycline binary system to study the effect of the timing of T-antigen expression on hyperplasia in the submandibular gland. Hyperplasia was reversed when T-antigen expression was silenced after 3 weeks; however, when T-antigen expression was silenced after 7 months, hyperplasia persisted even though the T-antigen was absent.
RU486 Binary System
O'Malley's laboratory developed a binary expression system that is regulated by RU486 (Wang and others 1994) and is based on the GAL4 expression system in yeast. The components of this system are shown in Figure 5B. In the RU486 binary system, the transgene of interest is fused to a genetically engineered promoter that contains the GAL4-bind-ing sequences. The chimeric transactivator contains the mutant human progesterone receptor, a portion of the yeast transcriptional activator GAL4, and the carboxy-terminal domain of the herpes simplex V16 gene. The mutant progesterone receptor cannot bind progesterone or other endogenous hormones; however, it will bind the progesterone antagonist RU486. The binding of RU486 to the mutated progesterone receptor induces the binding of the chimeric transactivator to GAL4 sequences. Because GAL4-activated genes are not known to exist in mammalian cells, the chimeric transactivator will bind to and activate only the target gene containing the GAL4-binding sequences in its promoter. A stable rat fibroblast cell line containing the RU486 binary system was implanted into rats, and Wang and others (1994) showed that the administration of RU486 to the rats induced the expression of a reporter gene at doses of RU486 that were significantly lower than that required for the antiglucocorticoid effects of RU486. However, no transgenic mice have been produced with the RU486 binary system.
CONCLUSION
The development of techniques to produce transgenic animals has been an extremely important advancement in science because it has provided investigators with a unique system for studying the role of individual genes both in normal physiological functions and in disease processes in the whole animal. Transgenic animals are a valuable resource in aging research because an investigator can study the potential effect of a specific gene (or process) on aging. With conventional research models, it is difficult, if not impossible, to determine the importance of a specific gene on aging because of the global nature of aging (that is, the alteration of a very large number of processes with increasing age). In Table 1 are listed several currently available transgenic mouse models that might be useful in studying aging.
Although transgenic animals provide investigators with a system for directly studying the role of a specific gene in aging or age-related disease processes, the complexity of whole animal studies still makes it difficult to establish cause-effect relationships between changes in the expression of a gene and phenotypical/physiological changes. For example, transgenic animals have been shown to compensate for changes in the expression in specific genes, especially during embryonic development, and the insertion of transgenes into the genome of a mouse (or ES cell) can also alter the expression of endogenous genes. This makes it difficult to demonstrate that phenotypical or physiological changes are the direct result of changes in the expression of the transgene or the mutated gene. Redundancy in physiological/biochemical systems (such as blood pressure and volume, antioxidant defense system, and neuroendrocrine system) also makes transgenic experiments difficult to interpret since altering only 1 component in such physiological systems may have very little impact on the transgenic animal physiologically because the redundancy of the system allows the cells/tissues to compensate for the alterations in the expression of the transgene.
In using transgenic mice for aging studies, it is extremely important to consider the strain of mouse used to produce the transgenic mice. Unfortunately, almost all the strains of mice currently used to produce transgenic mice have not been used in aging research and have not been characterized with respect to survival or pathological lesions that occur with age. Nevertheless, this limitation can be circumvented by using strains of inbred mice that are widely used in aging research. For example, transgene constructs can be injected into the fertilized eggs from inbred strains of mice to produce transgenic mice that overexpress a transgene. In addition, mice containing mutated genes can be backcrossed to inbred strains of mice producing congenic lines of the transgenic mice that are nearly identical to the parent inbred strain.
1Abbreviations used in this article: CMV, cytomegalovirus; ES, embryo-derived stein cells; Mn-SOD, Mn-superoxide dismutase.
ACKNOWLEDGMENTS
This article was supported in part through the Nathan Shock Center on Excellence in Basic Biology of Aging grant (PO3 AG13319) and the Nutritional Probe of the Aging Process grant (POI AG01188) from the National Institute on Aging and the Aging Research and Education Center at the University of Texas Health Science Center at San Antonio.
REFERENCES
Bailing R, Mutter G, Gruss P, Kessel M. 1989. Craniofacial abnormalities induced by ectopic expression of the homeobox gene Hox-I in transgenic mice. Cell 58:337-347.
Blackwell B, Bucci TJ, Hart RW, Turturro A. 1996. Longevity, body weight, and neoplasia in ad libitum-fed and diet restricted C57BL6 mice fed NIH-31 open formula diet. Toxicol Pathol 23:570-582.
Boulikas T. t993. Nature of DNA sequences at the attachment regions of genes to the nuclear matrix. J Cell Biochem 52:14-22.
Brinster RL, Allen JM, Behringer RR, Gelinas RE, Palmiter RD. 1988. Introns increase transcriptional efficiency in transgenic mice. Proc Natl Acad Sci U S A 85:836-840.
Brinster RL, Chen HY, Trumbauer ME, Yagle MK, Palmiter RD. 1985. Factors affecting the efficiency of introducing foreign DNA into mice by microinjecting eggs. Proc Natl Acad Sci U S A 82:4438-4442.
Cerami A. 1985. Hypothesis: Glucose as a mediator of aging. J Am Geriatr Soc 33:626-634.
Choi T, Huang M, Gorman C, Jaenisch R. 1991. A generic intron increases gene expression in transgenic mice. Mol Cell Biol 11:3070-3074. Coulson PS, Wilson RA. 1989. Portal shunting and resistance to Schistosoma mansoni in 129 strain mice. Parasitology 99:383-389.
Cruse JM, Lewis RE Jr. 1994. Transgenics today. Transgene 1:133-134.
Denlinger RH, Koestner A, Wechsler W. 1974. Induction of neurogenic tumors in C3HeB/FeJ mice by nitrosourea derivatives: Observations by light microscopy, tissue culture, and electron microscopy. Int J Cancer 13:559-571.
Dressler GR, Wilkinson JE, Rothenpieler UW, Patterson LT, Simons LW, Westphal H. 1993. Deregulation of Pax-2 expression in transgenic mice generates severe kidney abnormalities. Nature 362:65-67.
Duff K. 1997. Alzheimer transgenic mouse models come of age. TINS 20:279-280.
Duhamel-Clerin E, Villarroya H, Mehtali M, Lapie P, Besnard F, Gumpel M, Lachapelle F. 1994. Cellular expression of an HMGCR promoter-CAT fusion gene in transgenic mouse brain: Evidence for a developmental regulation in oligodendrocytes. Gila I 1:35-46.
Dycaico M J, Provost GS, Kretz PL, Ransom SL, Moores JC, Short JM. 1994. The use of shuttle vectors for mutation analysis in transgenic mice and rats. Mutation Res 307:461-478.
Efrat S, Fusco-DeMane D, Lemberg H, A1Emran O, Wang X. 1995. Conditional transformation of a pancreatic b-cell line derived from transgenic mice expressing a tetracycline-regulated oncogene. Proc Natl Acad Sci U S A 92:3576-3580.
Ewald D, Li M, Efrat S, Auer G, Wall R J, Furth PA, Hennighausen L. 1996. Time-sensitive reversal of hyperplasia in transgenic mice expressing SV40 T antigen. Science 273:1384-1386.
Finch CE. 1972. Enzyme activities, gene function, and aging in mammals. Exp Gerontol 7:53-67.
Furth PA, Hennighausen L, Baker C, Beatty B, Woychick R. 1991. The variability in activity of the universally expressed human cytomegalovirus immediate early gene I enhancer/promoter in transgenic mice. Nucleic Acids Res 19:6205-6208.
Furth PA, St. Onge L, Boger H, Gruss P, Gossen M, Kistner A, Bujard H, Hennighausen L. 1994. Temporal control of gene expression in transgenic mice by a tetracycline-responsive promoter. Proc Natl Acad Sci U S A 91:9302-9306.
Goodrich JA, Hoey T, Thut CJ, Admon A, Tjian R. 1993. Drosophila TAFii40 interacts with both a VPI6 activation domain and the basal transcription factor TFIIB. Cell 75:519-530.
Gossen M, Bujard H. 1992. Tight control of gene expression in mammalian cells by tetracycline-responsive promoters. Proc Natl Acad Sci U S A 89:5547-5551.
Hazzard DG. 1991. Relevance of the rodent model to human aging studies. Neurobiol Aging 12:645-649.
Hazzard DG, Bronson RT, McClearn GE, Strong R. 1992. Selection of an appropriate animal model to study aging processes with special emphasis on the use of rat strains. J Gerontol 47:B63-B64.
Hazzard DG, Soban J. 1988. Studies of aging using genetically defined rodents: A bibliography. Exp Aging Res 14:59-81.
Heynes NE, Kennedy N, Rahmsdorf U, Groner B. 1981. Hormone-responsive expression of an endogenous proviral gene of mammary tumor virus after molecular cloning and gene transfer into cultured cells. Proc Natl Acad Sci U S A 78:2038-2042.
Ho YS. 1994. Transgenic models for the study of lung biology and disease. Am J Physiol 266:L319-L353.
Hogan B, Beddington R, Costantini F, Lacy E. 1994. Manipulating the Mouse Embryo. Plainview: Cold Spring Harbor Laboratory Press.
Liu M, Olson AL, Moye R, Buse JB, Bell GI, Pessin JE. 1992. Expression and regulation of the human GLUT-4 muscle-fat facilitative glucose transporter gene in transgenic mice. J Biol Chem 267:11673-11676.
Kagan D, Platt J, Logan J, Byrne GW. 1994. Expression of complement regulatory factors using heterologous promoters in transgenic mice. Transplant Proc 26:1242.
Knapp JR, Kopchick JJ. 1994. The use of transgenic mice in nutrition research. J Nutr 124:461-468.
Koretsky AP. 1992. Investigation of cell physiology in the animal using transgenic technology. Am J Physiol 262:C261-C275.
Kothary R, Barten SC, Franz T, Norris ML, Hettle S, Surani MAH. 1991. Unusual cell specific expression of a major human cytomegalovirus immediate early gene promoter-lacZ hybrid gene in transgenic mouse embryos. Mech Ageing Devel 25-31.
Lebovitz RM, Zhang H, Vogel H, Cartwright J, Dionne L, Lu N, Huang S, Matzuk MM. 1996. Neurodegeneration, myocardial injury, and perinatal death in mitochondrial superoxide dismutase-deficient mice. Proc Natl Acad Sci U S A 93:9782-9787.
Lee F, Mulligan R, Berg P, Ringold G. 1991. Glucocorticoids regulate expression of dihydrofolate reductase cDNA in mouse mammary tumour virus chimaeric plasmids. Nature 294:228-232.
Li Y, Huang TT, Carlson EJ, Melov S, Ursell PC, Olson JL, Noble LJ, Yoshimura MP, Berger C, Chan PH, Wallace DC, Epstein CJ. 1995. Dilated cardiomyopathy and neonatal lethality in mutant mice lacking manganese superoxide dismutase. Nat Genet 11:376-381.
Masoro EJ. 1984. Food restriction and the aging process. J Am Geriatr Soc 32:296-300.
Masoro EJ. 1985. Nutrition and aging: A current assessment. J Nutr 115:842-848.
Masoro EJ. 1991. Use of rodents as models for the study of "normal aging": Conceptual and practical issues. Neurobiol Aging 12:639-643.
Mayo KE, Warren R, Palmiter RD. 1982. The mouse metallothionein-I gene is transcriptionally regulated by cadmium following transfection into human or mouse cells. Cell 29:99-108.
McKnight RA, Shama A, Sankaran L, Wall RJ, Hennighausen L. 1992. Matrix-attachment regions can impart position-independent regulation of a tissue-specific gene in transgenic mice. Proc Natl Acad Sci U S A 89:6943-6947.
Montoliu L. 1994. Transgenic mice: Fade to grey. Curr Biol 4:1025
Morimura S, Suzuki T, Hochi S, Yuki A, Nomura K, Kitagawa T, Nagatsu 1, Imagawa M, Muramatsu M. 1993. Trans-activation of glutathione transferase P gene during chemical hepatocarcinogenesis of the rat. Proc Natl Acad Sci U S A 90:2065-2068.
Muglia L, Jacobson L, Dikkes P, Majzoub JA. 1995. Corticotropin-releasing hormone deficiency reveals major fetal but not adult glucocorticoid need. Nature 373:427-432.
Nilsson E, Lendahl U. 1993. Transient expression of a human b-actin promoter/lacZ gene introduced into mouse embryos correlates with a low degree of methylation. Mol Reprod Dev 34:149-157.
Norwood VF, Gomez RA. 1994. Bridging the gap between physiology and molecular biology: New approaches to perpetual questions. Am J Physiol Regul lntegr Comp Physiol 267:R865-R878.
Oberly TD, Coursin DB, Cihla HP, Oberly LW. El-Sayyad N. Ho YS. 1993. lmmunolocalization of manganese superoxide dismutase in normal and transgenic mice expressing the human enzyme. Histochem J 25:267-279.
Olson EN, Arnold HH, Rigby PWJ, Wold BJ. 1996. Know your neighbors: Three phenotypes in null mutants of the myogenic bHLH gene MRF4. Cell 85:1-4.
Pahlavani MA, Haley-Zitlin V, Richardson A. 1994. Influence of dietary restriction on gene expression: Changes in the transcription of specific genes. In: Yu BP, editor. Modulation of the Aging Process by Dietary Restriction. Boca Raton, FL: CRC Press. p 143-156.
Palmiter RD, Sandgren EP, Avarbock MR, Allen DD, Brinster RL. 1991. Heterologous introns can enhance expression of transgenes in mice. Proc Natl Acad Sci U S A 88:478-482.
Passerman RS, Fishman GI. 1994. Regulated expression of foreign genes in vivo after germline transfer. J Clin Invest 94:2421-2425.
Peters J. Munter K. Bader M, Hackenthal E. Mullins JJ, Ganten D. 1993. Increased adrenal renin in transgenic hypertensive rats, TGR(mREN2)27, and its regulation by cAMP, angiotensin II, and calcium. J Clin Invest 91:742-747.
Richardson A, Birchenall-Sparks MC. 1983. Age-related changes in protein synthesis. In: Rothstein M, editor. Biological Research in Aging. New York: Arm R. Liss, Inc. p 255-273.
Richardson A, Roberts MS, Rutherford MS. 1985. Aging and gene expression. In: Rothstein M, editor. Review of Biological Research in Aging. New York: Alan R. Liss, Inc. p 395-419.
Richardson A, Semsei I. 1987. Effect of aging on translation and transcription. In: Rothstein M, editor. Review of Biological Research in Aging. New York: Alan R. Liss, Inc. p 467-483.
Rudnicki MA, Braun T, Hinuma S, Jaenisch R. 1992. Inactivation of MyoD in mice leads to up-regulation of the myogenic HLH gene M):/25 and results in apparently normal muscle development. Cell 71:383-390.
Rudnicki MA, Schnegelsberg PN, Stead RH, Braun T, Arnold HH, Jaenisch R. 1993. MyoD or MxJ:5 is required for the formation of skeletal muscle. Cell 75:1351 - 1359.
Rusconi S. 1991. Transgenic regulation in laboratory animals. Experientia 47:866-877.
Sands AT, Hansen TN, DeMayo FJ, Stanley LA, Xin L, Schwartz RJ. 1993. Cytoplasmic b-actin promoter produces germ cell and preimplantation embryonic transgene expression. Mol Biol Rep 34:117-126.
Schlager G. 1966. Systolic blood pressure in eight inbred strains of mice. Nature 212:519-520.
Schmidt EV, Christoph G, Zeller R, Leder P. 1990. The cytomegalovirus enhancer: A pan active control element in transgenic mice. Mol Cell Biol 4406-4411.
Smith GS, Walford RL, Mickey MR. 1973. Lifespan and incidence of cancer and other diseases in selected long-lived inbred mice and their Fl hybrids. J Natl Cancer Inst 50:1195-1213.
Sohal RS. 1993. The free radical hypothesis of aging: An appraisal of the current status. Aging Clin Exp Res 5:3-17.
Staats J. 1972. Standardized nomenclature for inbred strains of mice: Fifth listing. Cancer Res 32:1609-1646.
Staats J. 1981. Inbred and segregating inbred strains. In: Anonymous. Thc Mouse in Biomedical Research. New York: Academic Press, Inc. p 177-213.
StC. Sinclair NR. 1995. The trouble with transgenic mice. Immunol Cell Biol 73:169-173.
Stevens LC. 1973. A new inbred subline of mice (129/terSv) with a high incidence of spontaneous congenital testicular teratomas. J Natl Cancer Inst 50:235-242.
Stief A, Winter DM, Stratling WH, Sippel AE. 1989. A nuclear DNA attachment element mediates elevated and position-independent gene activity. Nature 341:343-345.
Sullivan N, Gatehouse D, Tweats D. 1993. Mutation, cancer and transgenic models: Relevance to the toxicology industry. Mutagenesis 8:167-174.
Swenberg JA, Koestner A, Wechsler W. 1972. The induction of tumors of the nervous system with intravenous methylnitrosourea. Lab Invest 26:74-85.
Talbot D, Descombes P, Schiber U. 1994. The 5'-flanking region of the rat LAP(C/EBPB) gene can direct high-level, position-independent, copy number-dependent expression in multiple tissues in transgenic mice. Nucleic Acids Res 22:756-766.
Thompson S, Clarke AR, Pow AM, Hooper ML, Melton DW. 1989. Germ line transmission and expression of a corrected HPRT gene product by gene targeting in embryonic stem cells. Cell 56:313-32 I.
Turrens JF, Boveris A. 1996. Generation of superoxide anion by the NADH dehydrogenase of bovine heart mitochondria. Biochem J 421-427.
Van Remmen H, Ward W, Sabia RV, Richardson A. 1995. Effect of age on gene expression and protein degradation. In: Masoro E J, editor. Handbook of Physiology: Aging. New York: Oxford University Press. p 171-234.
Wagner J, Thiele F, Ganten D. 1995. Transgenic animals as models for human disease. Clin Exp Hypertens 17:593-605.
Wang Y, O'Malley B Jr., Tsai ST, O'Malley BW. 1994. A regulatory system for use in gene transfer. Proc Natl Acad Sci U S A 91:8180-8184.
Wurm FM, Gwinn KA, Kingston RE. 1986. Inducible overexpression of the mouse c-myc protein in mammalian cells. Proc Natl Acad Sci U S A 83:5414-5418.
Yamashita T, Kasai N, Miyoshi I, Sasaki N, Maki K, Sakai M, Nishi S, Namioka S. 1993. High level expression of human alpha-fetoprotein in transgenic mice. Biochem Biophys Res Commun 191:715-720.
TABLE 1 Potential transgenic mouse models of interest for aging research
| GLUT4 (Hemizygous) Transgenic Mice (Liu and others 1992) | |
| Genetic manipulation | Transgenic mice were produced using the human minigene (11.5 kb) for the GLUT4 transporter. |
| Physiological changes | The transgenic mice express the transgene in adipose tissue, heart, and skeletal muscle and show lower plasma glucose levels (15 to 20% decrease compared with nontransgenic littermates). |
| Potential value in aging | One can test the Glycation Theory of Aging (Cerami 1985). The lowered plasma glucose levels would be predicted to lead to reduced glycation. If glycation reactions are important in aging, the transgenic mice should live longer and age more slowly. |
| Amyloid Transgenic Mice (Duff 1997) | |
| Genetic manipulation | Transgenic mice were produced with various gene constructs containing mutations in the gene coding for the amyloid precursor protein (APP). |
| Physiological Changes | The expression of the APP transgenes results in amyloid deposits in the brains of the transgenic mice. |
| Potential Value in Aging | One can study the role of amyloid deposits in the pathology of Alzheimer's disease. |
| CRH Homozygous Knockout Mice (Muglia and others 1995) | |
| Genetic Manipulation | The entire coding sequence of the pre-pro-corticosterone-releasing hormone (CRH) gene was deleted. |
| Physiological Changes | The CRH-/- mice show normal basal levels of corticosterone but fail to exhibit the diurnal increase in plasma corticosterone, and corticosterone levels do not increase with restraint stress or dietary restriction. |
| Potential Value in Aging | One can test the role of glucocorticoids in aging and the antiaging action of dietary restriction. |
| Sod2~/- Heterozygous Knockout Mice (Li and others 1995) | |
| Genetic Manipulation | The third exon of the Sod2 (Mn-superoxide dismutase [Mn-SOD]) gene was deleted. |
| Physiological Changes | The activity of Mn-SOD is reduced approximately 50% in tissues of the Sod2+/- mice. |
| Potential Value in Aging | One can test the Oxidative Stress Theory of Aging (Sohal 1993). The reduced activity of Mn-SOD would be predicted to lead to increased oxidative damage to various macromolecules. If the accumulation of oxidative damage plays a role in aging, the Sod2+/- mice should show an accelerated rate of aging. |
FIGURE 1 Summary of the method used in producing transgenic mice by the microinjection of exogenous DNA into the pronuclei of fertilized eggs.
FIGURE 2 Schematic representation of a chimeric transgene containing a full-length cDNA (under the regulation of a heterologous promoter) fused to DNA fragments containing an intron and a polyA cleavage site.
FIGURE 3 Description of the positive-negative selection technique used in the targeting of genes in mouse ES cells. In this example, the targeting sequence is constructed to contain exon 1, interrupted by the neomycin resistant gene (neor). The targeting construct contains the following selectable markers: neor within the targeting sequence and the herpes simplex virus thymidine kinase (HSV-tk) genes in the flanking sequences. When a double-reciprocal recombination event occurs in the ES cells, the cognate endogenous locus will be replaced with the targeting sequences, and this will result in the two HSV-tk genes being excised and the neor gene being inserted into the genome of the ES cells. This homologous recombination will make the cells resistant to the antibiotic G148 because of the presence of the neor gene and resistant to purine analogs (for example, ganciclovir [GANC]) because of the absence of the HSV-tk genes. This technique can be used for targeting any gene in mouse ES cells.
FIGURE 4 Generalized outline of binary expression systems used in the temporal regulation of gene expression in mammalian systems. The chimeric transactivator gene is shown at the top of the figure, and the arrow points to the protein product coded by the transactivator gene, which contains a DNA binding domain (shaded circle) and a transcriptional activating domain (open circle). The second component of the binary system is shown at the bottom of the figure. The gene of interest is placed under the control of a transactivator dependent promoter, that is, a promoter containing the unique DNA sequence that is recognized by the DNA binding domain of the chimeric transactivator protein. The arrow to the left shows an example of a binary system in which the temporal expression of the transgene is regulated by a repressor. For example, the transactivator protein spontaneously binds to the transactivator-dependent promoter and induces the expression of the transgene. A repressor inhibits the binding of the transactivator protein to the transactivator-dependent promoter and inhibits the expression of the transgene. The arrow to the right shows an example of the binary system in which the temporal expression of the transgene is regulated by an activator, that is, the transactivator protein will only bind to the transactivator-dependent promoter and initiate the transcription of the transgene in the presence of an activator.
FIGURE 5 A description of the 3 specific components of the tetracycline and RU486 binary expression systems. Panel A shows the following 3 components of the tetracycline binary system: (1) The chimeric transactivator consists of the E. coli tetracycline-repressor gene fused to the DNA sequence coding for the carboxy terminal domain of the herpes simplex V16 protein. The tetracycline-repressor domain of the chimeric transactivator will bind to the heptad sequences of the tetracycline operator, and the carboxy terminal domain of the herpes simplex V 16 gene will initiate transcription of the transgene. For example, this domain of herpes simplex V 16 contains a potent transactivation function that interacts directly with transcription factors TFIIB, TATA-binding protein, and TBP-associated factor TAFII40 within the transcriptional preinitiation complex (Goodrich and others 1993). (2) The transactivator-dependent promoter contains a heptad of the E. coli tetracycline operator sequence fused to a minimal human CMV promoter. The genetically engineered chimeric transactivator binds to the heptad sequence in the promoter. (3) Tetracycline acts as a repressor in this binary system. It binds with high affinity to the tetracycline-repressor domain, and this binding induces allosteric changes in the chimeric transactivator that prevents the transactivator from binding to the tetracycline operator.
Panel B shows the following 3 components of the RU486 binary system: (1) The chimeric transactivator contains the mutant human progesterone receptor, a portion of the yeast transcriptional activator GAL4, and the carboxy-terminal domain of the herpes simplex V16 gene. The mutant progesterone receptor, which contains a 42-amino acid deletion in its carboxy-terminal region, cannot bind progesterone or other endogenous hormones; however, it can bind RU486. The portion of the yeast GAL4 protein (residues 1-94) that is used in the chimeric transactivator contains a DNA-binding function, a dimerization function, and a nuclear localization signal. To enhance the transcriptional activity of the transactivator, the carboxy terminal domain of the herpes simplex VI6 gene is fused to the amino-terminus of the construct. (2) The transactivator-dependent promoter contains multiple copies of the yeast 17-mer GAL4-binding sequences fused to the minimal promoter of thymidine kinase. (3) RU486 acts as an activator in this binary system. It binds to the progesterone receptor domain of the chimeric transactivator, which induces the dimerization of the transactivator, and the binding of the GAL4 domain to the GAL4 sequences in the transactivator-dependent promoter.
Copyright © 2008. National Academy of Sciences.
All rights reserved.
500 Fifth St. N.W., Washington, D.C. 20001.
Terms of Use and Privacy Statement