The Mouse as a Model System


Introduction

     The mouse is a popular model system because it is a mammal with sophisticated genetic tools and significant genetic resources.

Position among mammals

     Mammalian orders arose from 50-65 million years ago, in a rapid diversification. The branching of the different mammalian orders was difficult to resolve on morphological and fossil evidence. Comparison of assembled whole genome sequences suggest that carnivores (dog) are more related to primates (humans) than rodents (mice).

Genome

     Mice have 20 chromosomes in their haploid genome (thus 40 chromosomes in all). The haploid genome is about 3 picograms, similar to humans. The gene order of the genomes of mice and humans are conserved (synteny) although there are rearrangements, several per chromosome. Unlike the mostly metacentric chromosomes of humans, all mouse chromosomes are acrocentric.

Physical

     Adult mice weigh 30-40 grams (50,000 to 70,000 grams for a young adult human) have a blood volume of 2 ml (4,800 ml for humans), and a resting heart rate of 500-700 bpm (60-80 bpm for humans).

Genetics

     Laboratory mice are unique in that there are a large number (hundreds) of inbred strains. Inbred strains are a strain of mice in which every individual is essentially genetically identical and homozygous at all loci.

     Inbred strains of mice are generated by 20 generations or more of brother-sister mating. Although it is often stated that inbred strains of mice are homozygous at all loci, minor variations within an inbred strain have been found, and clear differences exist between the same strain maintained at different vendors or laboratories.

     Most common inbred strains of mice arose from a limited number of genetically distinct mice, and thus the genome of many of the common inbred strains are a mosaic of small (several genes) contiguous stretches of one of a small number of genetic variants. Thus, for a span of several contiguous genes (300-600 kb), two strains may be essentially identical in DNA sequence (1 nucleotide change per 21,000 base pairs), or they may be divergent (1 nucleotide change per 440 base pairs). (Frazer et al., Nature 448:1050)

     The advantages of inbred strains of mice include the fixation of genetic background and the reproducibility of that background in different laboratories and through time (for some strains, like the common C57BL/6J from The Jackson Laboratory, the strain has been archived as frozen embryos and the stock is replace from frozen embryos periodically). Given the mutation rate (1 x 10^-5 per locus per gamete), genetic drift is low, and all mice of a given strain are essentially genetically identical (excepting males/females)

     When mice of two different inbred strains are mated, their offspring are said to be F1 (filial generation one) mice. F1 mice are genetically identical to each other, since each inherited the same paternal set of chromosomes and the same maternal set of chromosomes. F1 mice are more robust than their parents due to hybrid vigor. (Obviously, with the exception that males and females are chromosomally, genetically and phenotypically distinct. Also, there may be epigenetic parent-of-origin effects, such that F1's generated female strain1 x male strain2 could be different from F1's generated from female strain2 x male strain1--the two crosses in this example are called reciprocal crosses.)

     F1 mice mated brother-sister produce an F2 generation. All F2 mice are genetically distinct from each other, since the paternally and maternally rederived chromosomes recombine before segregating into gametes.

     An F1 mouse mated to its parent produces the first backcross generation, the N1 generation. Continued backcrosses to this strain generates subsequent backcross generations (N2, N3 etc).

     The terms isogenic, coisogenic, congenic and consomic are used to describe specific types of relatedness. Isogenic mice are genetically identical, thus different individual mice of an inbred strain are isogenic. Coisogenic mice have a variant (mutation, transgene, targeted allele) which arose directly on that strain. Congenic mice have a variant larger than a gene but are otherwise isogenic. Congenic mice typically were mutants which arose in one strain, but which were backcrossed 9 generations or more onto a second strain, selecting for the desired variant at each generation. The generations are denoted like a backcross with N1, N2, etc. In a congenic, the variant as well as the contiguous genes (from the first strain) near the variant are brought into the background since recombination will not efficiently remove closely linked DNA. Consomic mice have a complete chromosome from one strain of mice in the background of a second strain. A set of 21 consomic lines can be used to represent each (nuclear) chromosome of one strain in a second.

     There are also mice which are maintained as populations of genetically heterogeneous mice which are bred in rotation to maintain a high degree of heterozygosity. These strains are called outbred strains. Some outbred strains were put under selective pressure for high reproduction in their past. The robustness and high reproduction of the outbred strains make them useful in reproductive studies, reproductive technologies and in embryology, where large numbers of mutant embryos (which might be only 1/4, 1/8 or 1/16 of the embryos) are needed. The outbred strains of mice shouldn't be thought of as more representative of wild mice than inbred mice: both are laboratory artifacts, although generated by different breeding strategies and selective pressures.

     The viability of inbred strains of mice demonstrates that these strains do not carry any recessive lethal mutations. Recessive lethal mutations and other deleterious mutations are present in all wild animal populations (it has been estimated that each individual carries 6 recessive mutations), thus inbreeding results in reduced fitness and lethality. It must be assumed that lethal and deleterious mutations were present in the progenitors of laboratory mice, but lethals weren't propagated and deleterious mutations were selected against in a laboratory setting. Inbred strains do have genetic defects characteristic of the strain, including agenesis of corpus callosum (129 strain), blindness (C3H, FVB, SJL) and progressive hearing loss (A/J, Balb/c, C57BL/6).

     Mutations which affected the coat color of mice were among the first mutant alleles studied in the mouse, and they still are important today. Many inbred and outbred strains of mice carry a recessive mutant allele of a coat color gene (or combinations of recessive alleles). These recessive mutations were useful to detect contamination of an inbred strain with either wild mice, or another inbred strain.

     It is useful to understand the genetics of two of the most common coat color genes, albino and nonagouti. These genes are names for their recessive mutant phenotypes, so mice homozygous mutant at the albino locus have no pigmentation (white fur and pink eyes) and heterozygotes and wild type for albino have pigmentation (dark fur and eyes). The albino gene product, tyrosinase, is necessary for the synthesis of all pigment. Mice have two different pigments in each hair, yellow at the tip and black in the rest of the hair. In albino mutants, neither pigment is made. The nonagouti gene product is necessary for the production of the yellow pigment, so mutant nonagouti mice have black hair. In nonagouti heterozygotes and wild types, the fur looks brown because of the combination of yellow and black pigments. Mice mutant for both albino and nonagouti are white.

Life Cycle and Reproduction

     Human (and other higher primate) reproduction differs from the reproduction of most mammals, in that human reproduction is not seasonal and the ovulatory cycle is cryptic. Mouse reproduction in the wild varies seasonally, and the ovulatory cycle is evident and coordinates mating. Long daily light periods in the daily cycle support robust reproduction. Thus laboratory mice are maintained on a long (12 hours of light 12 hours of dark) light cycle.

     The ovulatory or estrus cycle of female mice has a length of 4 to 6 days. The estrus cycle can become synchronized in female mice housed together continuously, and can be suspended in the absence of exposure to male pheromones. Suspended cycles can be restarted by exposure to male pheromones. Female mice are only receptive to mating when they have ovulated. Receptive females can be identified by inspection of the genitalia. The physical stimulation of the cervix in mating is required to make the uterus receptive to embryo implantation. The male produces a copulatory plug from secretions from the vesicular and coagulating glands that blocks the vulva for 8 to 10 hours after mating. Whether a female has mated can be determined by the plug, by inspection or with a blunt probe.

     Estrus and ovulation can be manipulated by hormone injections to synchronize and maximize ovulation (superovulation). The early embryo (preimplantation embryo) is free in the oviduct and uterus of the female mouse for the first 4 days. Preimplantation embryos can be recovered from the oviduct and uterus by flushing culture media through the oviducts and uterine horns (the uterus of mice is bicornuate, having two tracts attached at the cervix.). Preimplantation embryos can be manipulated in culture by injection, infection and by adding or removing cells. They can be cultured and transferred surgically into a recipient female (embryo transfer). If the uterus of the recipient female is receptive, then pups can be born. Recipient females are mated to vasectomized males prior to the embryo transfer surgery to ensure that the uterus is receptive. These pseudopregnant females are mated according to a schedule such that the age of the transferred embryos is matched.

     Gestation in the mouse takes 18 to 20 days, depending on the strain. Within a litter of embryos, there is some variation in the timing of development, which is particularly evident at early stages when morphological development is rapid.

     The stages of embryos and fetuses are designated by a standardized nomenclature. This system expresses the stage of an embryo as the day of gestation. The midnight preceding the plug is designated as 0.0 days or E0.0. However, because individual embryos within a litter and embryos of different strains develop at different rates, the stage of an embryo, designated in days, is defined morphologically, in fact. For example, at eight days (midnight at the end of the eighth day), some embryos will have formed the first somite and will be E8.0, and others in a litter may be younger with well developed headfolds and be E7.75 and other might be older with 3 somites and be E8.25. Yes, it really is as confusing as you think it is.

     Most litters which are born can be successfully reared by their mothers. Pups which are not viable may be partially or wholly consumed by mice in the cage. When the mother is not happy, however, she may eat the entire litter. A key component of keeping mice happy is to keep them in stable social groupings. Thus, male-female pairs, pairs of acquainted pregnant females, or a non-pregnant and pregnant female are all more likely to successfully rear offspring than individually housed females.

     Mice are born hairless and with their eyes closed. By three weeks of age, they have their adult hair, open eyes, teeth and can jump to the top of the cage to feed and drink. At this time they can be removed from their mothers, and are typically housed together by sex.

     For targeted and transgenic strains, mice carrying the DNA alteration are usually identified by taking a punch of tissue from the ear in a pattern to mark the animal and to provide DNA for a PCR assay (genotyping). Genotyping is most often performed at weaning.

     Sexual maturity can occur as early as 6 weeks in females and 8 weeks in males, depending on strain.

     The reproductive life span varies widely with strain, but females typically have a reproductive life span of 6-8 months, whereas males have a reproductive life span of about a year. In practice, it is important not to overestimate reproductive life span in order not to lose the ability to propagate the strain.

     The life span of mice also varies with strain, but is 1.5-2 years typically.

Infectious Agents and Commensal Bacteria

     In addition to standardized genetic background, much effort is put into standardizing the pathogen status of mice. It seems obvious that pathogens and cycles of infection could impact research results, but the standard of practice goes beyond pathogens with clinical effects to include some agents without clinical effects in normal mice. The rationale for elimination of subpathogenic agents is that they may have immunomodulatory effects, since mice are widely used for immunological research. This standard of care is called Specific Pathogen-Free. This standard has changed as new pathogens are discovered or their effects are better understood. For example, noroviruses and Helicobacter sp were added to the list of excluded pathogens recently.

     Mice which have pathogens are made Specific Pathogen-Free by rederivation. Rederivation is an embryo transfer where the embryo donor mother is kept in quarantine, the embryos recovered sterilely and transported out of quarantine and transferred into a Specific Pathogen-Free pseudopregnant recipient female, housed in a facility which is Specific Pathogen-Free.

     Commensal bacteria can affect physiology and interact with genetic variants in the host. In order to control the commensals, mice free of bacteria (axenic mice) can be generated by caesarian delivery. Axenic mice can then be inoculated with defined bacterial populations to generate gnotobiotic mice.

Cell Culture

     Primary cultures from mouse embryos or fetuses (Primary Mouse Embryonic Fibroblasts or MEFs) can be used to look at cellular phenotypes. MEFs can be immortalized by extended culture until they become transformed, or cell lines from specific tissues can be generated by expression of SV40 T antigen in the mouse in the tissue of interest.

     Cell lines from preimplantation embryos can be generated. These embryonic stem cell lines (ES cells) can be used to introduce exogenous DNA and to alter endogenous DNA by gene targeting. These ES cells can be studied directly in culture or can be used to generate mice containing the genetic alterations. Existing mouse strains can also be used to generate new ES cell lines, also used for studying cellular phenotypes in vitro. ES cells have many advantages over MEFs, including more rapid proliferation and the formation of clonal colonies.

Chimeras

     Preimplantation embryos can be manipulated to incorporate other cells, and if the added cells are embryo-like (for example ES cells), they will participate in the formation of an embryo, fetus, and live born animal. ES cells can participate in the formation of all adult tissues, including the germ line, the cells which produce gametes. These embryos, fetuses and adults which are mixtures of two genetically distinct cells are called chimeras.

     Because the genomes of the different components of a chimera are in separate cells, the genomes do not recombine. Thus, if the host embryo is a/a and the ES cell is A/A, then both 'a' and 'A' sperm could be produced. But 'a' sperm must come from the host, and 'A' sperm must come from the ES cell. If this chimera is mated to a mouse which is a/a, then a/a offspring come from the host cells, and A/a offspring come from the ES cell.

     In the preceding example, the a and A are symbols used for the nonagouti locus, where a/a mice are black, and A/a mice (and A/A mice) are brown. This scheme is the strategy most often used to determine if the ES cells transmitted through the germ line into offspring.

     In a typical targeting experiment, only one of the two copies of the gene are targeted, so that only 50% of the brown offspring will carry the desired mutation.

     ES cells are typically made from males. The host embryos are not selected for sex, so chimeras can be made with a male host embryo or a female host. Because of the dominant nature of testosterone, and the fact that it acts systemically, the male/female chimeras will be male mice.

     Chimeras can also be used experimentally to examine the effect of a mutant gene on cells in the context of wild type tissue. This strategy can be used to determine if the effect of the mutant gene acts on the cell itself (autonomous) or on its adjacent or distant neighbors (nonautonomous).

Transgenic Mice

     Mice can be engineered to have an introduced gene, so-called transgenic mice. Typically, fertilized eggs (zygotes) are collected and microinjected with DNA. The DNA is injected into the sperm nucleus after fertilization but before it fuses with the egg nucleus--before fusion the haploid nuclei are called pronuclei, thus the injection is sometimes called pronuclear injection. Around 20% of the surviving injected eggs will have the DNA inserted at one site in a chromosome. The injected DNA typically inserts as multiple head-to-tail copies (10 to 100's of copies) at this one site.

     The insertion of DNA by microinjection often (~10% of the time) results in mutation. This high rate of mutagenesis occurs not because 10% of the genome is functional, but because the insertion of DNA is often accompanied by deletion and rearrangement at the site of insertion.

     Introduced genes most often are minigenes consisting of enhancer elements for tissue-specific expression, a transcriptional promoter, the coding sequences (cDNA) and a complete set of poly-adenylation signals.

     Surprisingly, the expression level of a minigene transgene does not correlate with the number of copies.

     The expression of transgenes can be lost in subsequent generations through epigenetic silencing, thus a means to monitor that the transgene continues to be expressed is important, and monitoring transmission of the DNA into offspring by itself is not enough.

     The expression of transgenes can be altered by the chromosomal environment around them. These effects include level of expression, tissues expressed in, and propensity to be silenced.

     Because not all transgenes are expressed as desired, and because the insertion may have disrupted an endogenous gene, multiple (~3) independent transgenic founder mice or lines should be analyzed to ensure that the phenotypes generated are due to the transgene. To facilitate this, each DNA-positive founder mouse from the injection is typically bred to wild type mice to establish a line of mice descending from that founder, and then the different founder lines are treated as independent experiments.

     Systems for temporal control of transgene expression also exist. One of the most prevalent systems is the doxycycline/tetracycline system. The strategy involves control of gene expression using the administration of a drug. Typically, these systems involve two transgenes, one which expresses a transcription factor which binds to the drug which activates or inhibits the transcription factor; the second has DNA sequences for binding of the transcription factor which regulate expression of the target sequences in cis.

     The cloning and tiling of the genome with large (~200 kb) genomic fragments in bacterial artificial chromosome (BAC) libraries has made possible the use of these fragments as transgenes. Unlike minigene transgenics, these large fragments direct expression proportionate to their copy number in the genome, often in the expression pattern of the endogenous gene and they are much less susceptible to epigenetic silencing. In addition, technologies are available (e.g., recombineering) to modify the DNA in sophisticated ways.

Gene Targeting

     The genes of ES cells in culture can be manipulated by homologous recombination. The ability to grow large number of cells permits the selection of cells with rare homologous recombination events, and the ability of ES cells to contribute to mice makes it possible to introduce these genetic alterations into mice, once the cells with rare events have been identified. The clonal growth of ES cells is important here again, in allowing the facile establishment of many individual lines of cells.

     Most often, gene targeting is used to generate a mutation such that the gene no longer produces a functional product (a null allele of the gene). A null allele is the most important genetic tool for determining the normal function of a gene. These mutations are colloquially referred to as knockouts.

     Gene targeting can used to introduce point mutations or other subtle sequence alterations or to put a complete new coding sequence into a gene. These mutations are colloquially referred to as knockins.

     Because a gene may have multiple functions at different times and in different tissues, it is useful to be able to eliminate gene function at specific times or in specific tissues. These alleles are termed conditional alleles, or, colloquially, floxed (for flanked by loxP) alleles. Conditional alleles generated by homologous recombination have wild type function until exposed to Cre recombinase, which deletes a portion of the gene which was flanked by the DNA sequences which bind Cre, the loxP sequence. In cases where Cre coding sequences are fused to a mutant form of the ligand-binding domain of the estrogen receptor (Cre-ER), the drug tamoxifen can be administered to mice to activate Cre-ER, initiating a round of recombination.

Gene Traps

     Gene trap libraries of mutated genes in ES cells are an important genetic resource. Gene trap libraries are generated by the random insertion of gene trap vector DNA into genes, with subsequent molecular identification of the gene. Gene trap vectors are typically one of two types, splice acceptor traps or poly-A traps. In a splice acceptor trap, when the gene trap inserts into an intron in the correct orientation, the endogenous gene's splicing pattern is disrupted by splicing into the splice acceptor of the gene trap. Splicing into the gene trap also results in expression of a drug resistance gene, and the survival of such cells in the presence of the drug. In a poly-A trap, the gene trap vector has a promoter and coding sequences for a drug resistance gene, but no polyadenylation signals. In the absence of a polyadenylation signal, not enough of the drug resistance protein is made to allow survival. If a poly-A trap vector inserts upstream of the poly-adenylation signals of a gene in the correct orientation, then properly terminated and adenylated transcripts for the drug resistance gene will be made, along with the drug resistance protein. To disrupt function of the gene, poly-A traps also have a splice acceptor to disrupt splicing of the endogenous gene. The genes which have been trapped are identified by sequencing of cDNAs spliced into gene trap sequences. The collections of gene traps are now large enough that genes with introns have been trapped multiple times. Gene traps with insertions in introns toward the 5' end of the gene are more likely to result in complete loss of function (null).

Mutant Resources and Repositories

     Scientific agencies world wide have recognized that their support is necessary for the preservation and distribution of variant mice. Mouse Genome Informatics (MGI) at The Jackson Laboratory maintains a database of mutant, transgenic and other variant mice which have been published. This is the place to start to determine which alleles of a gene have been published. MGI links to the International Mouse Strain Resource (IMSR), which is a database of mutants in public repositories as mice or cryopreserved gametes, embryos or ES cells. Alleles of genes in gene trap libraries are not currently accessible through MGI and IMSR. The two places to start your search for gene trap alleles are the International Genetrap Consortium and the Texas Institute for Genomic Medicine. Conditional gene traps also exist, and these can most easily be searched at the European Conditional Mouse Mutagenesis (EUCOMM) Program. Links to consortia working from the phenotypes of variants and induced mutants back to the mutated gene--forward or classical genetics--are available at MGI's Phenotypes and Mutants Community Resources page.

DNA Resources

     The Jackson Laboratory provides genomic DNA from many of their strains.

     BAC libraries of genomic DNA have been made from a number of inbred mouse strains. Two of these libraries, from 129 and C57BL/6J, have been end-sequenced and tiled on the assembled genome, so that the contents of the clones are known. For 129 genomic clones, BACs containing your gene can be identified using the ensembl genome browser, selecting the DAS source "129S7/AB2.2 clones". If you wish to obtain a specific clone, clicking on the BAC will bring up a menu, and selecting the link at the bottom of the list will take you to the order form to purchase from the Sanger Institute. C57BL6J BAC clones can be identified with the UC Santa Cruz genome browser and purchased from BACPAC Resources Center CHORI. The locations of single nucleotide polymorphisms between these strains can be ascertained here on the Jax web site.