Online Issues

Journal Vol 47(2)

Genetic Variables That Influence Phenotype

Carol Cutler Linder

Carol Cutler Linder, Ph.D., is currently an Assistant Professor of Biology, Department of Natural Sciences, New Mexico Highlands University, Las Vegas, New Mexico. Dr. Linder also was a Research Scientist at The Jackson Laboratory, Bar Harbor, Maine, during the preparation of this manuscript.

Abstract

Characterization of genetically engineered mice requires consideration of the gene of interest and the genetic background on which the mutation is maintained. A fundamental prerequisite to deciphering the genetic factors that influence the phenotype of a mutant mouse is an understanding of genetic nomenclature. Mutations and transgenes are often maintained on segregating or mixed backgrounds of often-unspecified origin. Minimizing the importance of strain and substrain differences, especially among 129 strains, can lead to poor experimental design or faulty interpretations of data. Genetic factors that influence phenotype can be categorized as traits that are unique to the background strain, unique to the gene of interest, or an interaction of both the background strain and the gene of interest. The commonly used inbred strains are generally well characterized and understood; however, specific genetic alterations combined with genes unique to the background inbred strain may lead to unexpected results. Genetic background effects can be analyzed and controlled for by using specific targeting and breeding strategies. Selection of appropriate experimental controls is critical. Ideally, mutations or transgenes should be characterized on more than one genetic background and in hybrids of the two progenitor strains. This approach may lead to the identification of novel genetic modifiers of the “gene of interest.” Conditional mutagenesis technologies increase the options for controlling genetic background effects in addition to permitting the study of developmental and temporal changes in gene and protein expression and thus phenotype.

Key Words: controls; genetic background; genetic modifiers; genetically engineered mice; inbred strains; knockouts; nomenclature; transgenics

Dr. Musmusculus is on top of the world—one of his knockout mice has a phenotype! The Dr. has invested 2 yr in gene targeting and chimeric mouse generation, and then another year in expanding and aging his knockout colony. Still, the Dr. must answer some difficult questions. Is the phenotype caused by the targeted gene? Is it a background effect? Is it premature to think about that elusive Nature Genetics paper?

Attributing an aberrant phenotype to one's current “gene of interest” is natural, especially after the years of time and money spent to create a genetically engineered mouse strain. In addition to an overexpressed or mutated gene, however, a number of genetic and environmental factors can contribute to the phenotypic characteristics of a mouse. Genetic factors that influence a mouse phenotype are the focus of this article. The attributes expressed in a mutant mouse are the result of genes present in the background strain working independently or in concert with the effects exerted by the gene of interest.

Mouse Nomenclature—Is It All in the Name?

It is important to understand the role of correct strain and gene nomenclature as a prerequisite to the exploration of genetic factors. Mouse nomenclature can be confusing and overwhelming to even the experienced mouse geneticist. A standardized nomenclature system is crucial, however, because mistakes or omissions can cost both time and money. A complete strain name is a unique identifier that provides information about the background strain, its source (i.e., vendor), and the official gene designation. Strain names allow for greater ease of access, comparison, and interpretation of data available through online databases and the published literature. Common mistakes and pitfalls associated with strain nomenclature issues that affect phenotypic evaluation are highlighted in the following paragraphs. Additional resources are provided in Table 1 to guide readers through the most complex nomenclature problems.

Strain nomenclature is often long and awkward; however, abbreviations (or shortcuts) provide incomplete information about a strain. Although it is acceptable to use appropriate abbreviations on cage cards and in laboratory notebooks when referring to a strain, full strain nomenclature should be used when ordering mice, when providing documentation in the literature (in Materials and Methods, Abstract, and Key Words sections of publications), and in permanent breeding and facility records. An appropriate abbreviation should include designations for the background strain and the particular allele of interest. Alleles are sequence variations in a gene that can occur spontaneously over time or be induced artificially. A targeted mutation (tm¹) is designated by appending an allele designation to the gene symbol. For example, Il4^tm1Cgn is the allele designation for a tm generated in the interleukin 4 (Il4) gene by Dr. Ralf Kuhn at the University of Cologne (Cgn). All too often, cage cards in an animal research facility omit such critical information. It is not uncommon to find mice designated simply as “Il4+/−.” Missing from the cage card is any indication of the background strain and a complete gene designation. According to the Mouse Genome Database (2005), there are 11 different targeted alleles of the Il4 gene, maintained on a variety of mixed or inbred backgrounds. Each allele represents a separate gene-targeting event potentially producing a different phenotype and requires a different genotyping protocol to identify mutant and heterozygous carriers.

Incomplete informal abbreviations are often used to designate strain names. For example, B6 is a common abbreviation for C57BL/6 mice, yet B6 does not specify the substrain designation and thereby omits critical source information. C57BL/6N mice are from the production facilities of the National Institutes of Health (NIH¹), whereas C57BL/6NCrl mice are from Charles River Laboratories (which originally obtained their substrain from the NIH). Each of the commonly used inbred strains such as C57BL/6, BALB/c, C3H, DBA/2, and 129 have multiple substrains and can be obtained from major mouse vendors as well as from in-house research mouse colonies around the world (Table 2). The characteristics of a mouse and subsequently its utility for a specific research application can vary depending on the particular substrain. Due attention and diligence to nomenclature will ensure that the correct mice are mated and/or selected, even when animal facility or research personnel change during the course of a project.

Standard formal strain abbreviations are used to describe mice with mutations maintained on a mixed genetic background. Targeted mutations created in a 129-derived embryonic stem (ES¹) cell line often are crossed to C57BL/6 mice to test for germ line transmission. C57BL/6 x 129 F2 mice (B6129F2) that carry the targeted mutation are then intercrossed to produce homozygous mutant mice. Continued breeding of the progeny of the F2 mice will generate a new “mixed inbred” strain after 20 generations of inbreeding (F20). Incipient inbred stocks (<F20) or inbred strains that are derived from only two parental strains are designated with uppercase abbreviations separated by a semicolon (e.g., B6;129-Il4^tm1Cgn). The semicolon designation for mixed inbred strains should not be confused with the period (.), which denotes congenic nomenclature. The period separates the host background strain from the donor strain (e.g., B6.129-Il4^tm1Cgn). The genetic background of mutations or transgenes maintained by backcrossing to an F1 hybrid (e.g., C57BL/6J x C3HeB/FeJ)F1, (C57BL/6J x SJL/J)F1) are designated by strain abbreviations with no punctuation (e.g., B6C3Fe-Tg, B6SJL-Tg). A mutation maintained on a genetic background derived from more than two strains, or from an unknown source, is considered a “mixed” inbred and may be designated as STOCK (International Committee on Standardized Nomenclature for Mice 2005).

A final consideration when coping with mouse nomenclature, strain selection, and experimental design is that proper and complete strain names are not a substitute for research in the scientific literature and a detailed understanding of the mutant gene of interest and the background strain on which it is maintained. The mouse community remains divided about the amount of information that should be conveyed via the gene symbol. For example, targeted genetic mutations are designated as separate alleles by appending tm, a sequential number, and the researcher's laboratory code (e.g., Il4^tm1Cgn,IL4^tm2Cgn, Il4^tm3Cgn). Although these mutations are commonly called knockouts, they may or may not completely eliminate protein function by design or default. It is important to read the primary reference for a specific mutation to determine the nature of the gene disruption, to obtain an allele-specific genotyping protocol, and to determine whether a functional or altered protein is produced.

In the absence of proper research, transgenic nomenclature can be just as confusing as that of targeted mutations. Expression of a transgene depends on the promoter construct and the site of integration. Because transgenic nomenclature does not always indicate the promoter used, it is imperative to consult the relevant literature to determine whether a transgenic strain is appropriate for an experiment.

Genetic Variables—What Causes a Phenotype?

A phenotype encompasses all of the observable traits of a mouse, including physical and behavioral characteristics, clinical profiles, histopathological features, as well as molecular characteristics (e.g., gene and protein expression profiles). A phenotype is a manifestation of a genotype, the mouse's genetic composition, in the context of environmental or epigenetic influences affecting gene expression. The individual genotype at a specific location (locus, allele) for genes on autosomal chromosomes will be either homozygous (with two identical alleles or variants, A/A or a/a) or heterozygous (with two different alleles, A/a).

The expression of a phenotype in mice carrying a spontaneous or induced gene mutation depends on a number of factors not readily apparent either to the researcher who generated the mice or to those using the model in subsequent studies. Dr. Barthold's Muromics review (Barthold 2002) provides a comprehensive analysis of genomic issues relevant to the use of laboratory mice. Traits expressed in a mouse arise from gene expression that is unique to the background strain, unique to the gene of interest, or an interaction of both the background strain and the gene of interest. Although some phenotypes due to specific mutation are found in all genetic backgrounds, phenotypic variability often becomes apparent only when a mutation is studied on numerous different genetic backgrounds. This type of modulation in a mutant phenotype is typically referred to as a genetic background effect. Another factor that contributes to variable and/or unexpected phenotypes observed in mutant mice is the inadvertent introduction of a hypomorphic, rather than a null, mutation. A hypomorphic mutant gene displays a partial, rather than complete, reduction in the activity it influences. Alternately, there may be compensatory pathways in the mouse that are upregulated in the absence of a normally expressed gene.

Transgenic strains may exhibit variable expression as a result of the chromosomal integration site, transgene copy number, or long-term chemical modifications to the incorporated transgene. The transgene integration site may affect a phenotype on multiple levels (see Tinkle and Jay 2002 for an overview). There could be more than one transgene integration site within a single founder line; multiple integrations sites should be separated by breeding into two independent lines. In addition, different transgenic founder lines have different sites of integration; it is important to characterize the phenotype in mice from several founder lines to determine whether the resulting phenotype is linked to the transgene insertion site. Finally, there may be variegated expression among siblings from the same founder. Insertion into areas adjacent to constitutive heterochromatin may lead to epigenetic factors that influence transgene expression, for which BLG/7 transgenic mice that express an ovine β-lactoglobulin transgene are an example (Opsahl et al. 2002).

Transgenes usually integrate multiple copies into the genome in a head-to-tail orientation (Tinkle and Jay 2002). Transgene copy number often affects the resulting phenotype and should be considered when establishing a breeding scheme because generating a homozygous colony effectively doubles the transgene copy number. Transgene copy number is not always stable, as demonstrated by the transgene carrying a human Cu,Zn superoxide dismutase mutation (SOD1G93A) created by Dr. Mark Gurney and colleagues (1994). A recombination event occurred in the strain during establishment of a breeding colony at The Jackson Laboratory (Gurney 1997); additional recombination events have been observed in different laboratories (Alexander et al. 2004; Gurney 1997). The onset of the neurodegeneration in this model of amyotrophic lateral sclerosis correlates very well with transgene copy number (Alexander et al. 2004; Gurney 1997). Lastly, transgene expression is regulated by epigenetic factors like DNA methylation and chromatin modifications (Chevalier-Mariette et al. 2003; Padjen et al. 2005).

Genetic background is defined as a collection of all genes present in an organism that can influence a trait or traits. Although most of the commonly used inbred strains share a relatively common origin (Beck et al. 2000; Silver 1995b), each strain has its own unique set of characteristics (Table 2). Strain characteristics are the result of allelic variation among inbred strains. Parent strains are further divided into substrains, many of which have been separated for numerous decades. Related strains are independently exposed to genetic drift, retrovirally induced recombination insertions and transpositions, and accidental or deliberate outcrossing. Strain attributes can be the result of single genes (e.g., a retinal degeneration mutation that causes blindness in inbred strains like C3H/HeJ, FVB/N, and SJL/J) or a combination of genes. Examples of strain-specific polygenic traits include differential susceptibilities of inbred strains to development of diet-induced obesity (West et al. 1992) and atherosclerosis (Nishina et al. 1993) and differences in reproductive potential (Canning et al. 2003; Festing 2001). Strains for mutant gene characterization are often selected without investigation into the characteristics of the parent strain or the identification of the presence of alleles that will invalidate experimental results. For example, common inbred strains can carry alleles causing callosal agenesis, blindness, and hearing loss (Table 2).

Selection of a 129 strain to match genetic mutations created in 129-derived embryonic stem (ES¹) cell lines is particularly challenging. Genetic and phenotypic analysis of the 129 mice available to the research community revealed occurrences of genetic contamination and deliberate cross breeding, resulting in significant genetic diversity (Simpson et al. 1997; Threadgill et al. 1997). Functional differences among the different 129 strains also exist. Although all 129 mice express the same major histocompatibility complex haplotype (H2^b), unidentified differences in minor histocompatibility genes are sufficient to cause tail skin graft rejection between most 129 strain lineages (Simpson et al. 1997). Substrains within the 129P and 129X parental lines carry recessive mutations in the tyrosinase (Tyr) and pink-eyed dilution (p) genes, which impair vision and potentially confound spatial learning and memory tests. Given the importance of understanding the origin of 129 mice, their nomenclature was modified to identify separate families derived from common parent strains (Festing et al. 1999). Researchers must exercise caution in choosing a 129 strain for breeding purposes inasmuch as the published nomenclature is often incomplete or misleading. Many ES cell lines are derived from strains within the 129S lineage (Simpson et al. 1997), yet the strain origin is generally referred to simply as “129/Sv.” This type of incomplete information may lead researchers to conclude erroneously that 129/SvJ (now 129X1/SvJ) mice are a suitable match. In fact, although there are ES cell lines derived from 129X1/SvJ, it is the most genetically distinct 129 strain and is not a good match to 129S-derived ES cell lines (Simpson et al. 1997; Threadgill et al. 1997).

Genetic modifiers are alleles present in the background strain genome that alter the expression of the gene of interest. Genetic modifiers may function through one of numerous mechanisms, including (1) suppression or enhancement of expression of genes involved in physiological or pathological pathways, (2) alteration of DNA transcription rates or mRNA stability, (3) epigenetic effects causing changes in DNA methylation or chromatin structure, and (4) variation of gene copy number (Erickson 1996). Traditional gene mapping and positional cloning approaches are used to determine the chromosomal location and to identify specific genetic modifiers (Davisson 2006; Montagutelli 2000; Silver 1995a). Mapping begins with F2 intercrosses between two strains that harbor the mutation but display differing phenotypes. Phenotypic screening of F2 mice is followed by a genome-wide analysis using DNA markers that are polymorphic between the two strains.

Regulation of tumor incidence in mice heterozygous for the chemically induced multiple intestinal neoplasia mutation (Apc^Min) provides an example of genetic modifiers. The modifier of Min 1 (Mom1) confers resistance to intestinal tumorigenesis. Mom1 was mapped to chromosome (Chr¹) 4 by linkage analysis; subsequent analysis identified secretory type II phospholipase 2A (Pla2g2a) as the candidate gene (Cormier et al. 2000). Tumor incidence is greater in C57BL/6J-Apc^Min mice, which are homozygous for a null allele of Pla2g2a,compared with AKR/J-Apc^Min mice, which are wild-type for Pla2g2a . Additional modifiers map to Chr 18 and cause significant alterations in tumor multiplicity (Haines et al. 2005; Silverman et al. 2002).

The use of inbred strains for the characterization of modified genes, including spontaneously occurring or induced mutations, is ideal because inbred strains allow the characterization of a specific genetic alteration on a uniform genetic background. Interpreting phenotypic information on a mixed segregating genetic background can be problematic. Underlying genetic complexities are illustrated by the use of targeted disruptions in the insulin receptor (Insr) to examine mechanisms of insulin resistance in type 2 diabetes. Neonatal lethality due to severe diabetes is observed in homozygous null mice on all genetic backgrounds. Variable hyperinsulinemia is observed in heterozygous mice on a mixed segregating background, indicating phenotypic interference by undefined background modifiers (Accili et al. 1996). However, on a C57BL/6 congenic background, heterozygous insulin receptor knockouts show only mild hyperinsulinemia. In contrast, heterozygous mice on a 129S6 congenic background show severe hyperinsulinemia and insulin resistance. Five quantitative trait loci (QTL¹) contributing to the phenotype were identified in progeny of an intercross between C57BL/6 and 129S6 congenic stocks. Of these, four deleterious QTL were contributed by the seemingly hyperinsulinemia-resistant C57BL/6 background, with only one deleterious QTL identified from the hyperinsulinemia-permissive 129S6 background (Kido et al. 2000).

An observed phenotype may be due to genes completely independent of the modified gene or transgene. During congenic strain development, a gene linked to a modified gene may be carried over during backcrossing and contribute to the phenotype of the mouse. Linked genes affecting coat color become evident as soon as mice are intercrossed to produce homozygotes. Pressures to publish work involving newly created mice combined with the time involved to create fully backcrossed congenic strains, defined as 10 generations of backcrossing (N10), often leads to the premature characterization of a background that is still segregating for genes from progenitor strains.

Targeted genes linked to coat color alleles that differ between the ES cell line and the host background may cause a coat color alteration even in a fully congenic C57BL/6 strain. For example, the 129-derived white-bellied agouti allele is linked to Rag1 on Chr 2. The 129 mice carrying the Rag1 targeted mutation backcrossed to C57BL/6 remained white-bellied agouti (Figure 1). Although alarming to the animal technician or researcher, a tightly linked gene may provide a shortcut to genotyping mutant mice. Ideally, the congenic interval should be minimized by further backcrossing and the use of polymorphic DNA markers to avoid misattribution of a phenotype to the modified gene.

Figure 1. The Rag1 gene is closely linked to the white-bellied agouti allele (A^w) on chromosome 2. The Rag1 targeted mutation (Rag1^tm1Tyj) was made in 129S1-derived embryonic stem cells and then backcrossed onto a C57BL/6 background. Fully congenic mice at N10 were still segregating for coat color. Some mice were white-bellied agouti because the congenic interval included the A^w allele from 129S1 (left), while others were nonagouti black (a) because a recombination event occurred between the agouti and Rag1 loci leading to a shortened congenic interval that does not include the A^w allele (right). Black siblings were selected for mating to establish a nonsegregating colony.

Several viable solutions have been used to eliminate flanking gene effects on the mutant gene of interest. A coisogenic strain can be created by using a DNA library, an ES line, and a background strain all derived from the same inbred strain. This strategy is theoretically attainable in 129 mice; however, 129 may not be the wisest strain choice. Although most widely used ES cell lines are derived from 129, genetic diversity among substrains creates specific problems (Simpson et al. 1997; Threadgill et al. 1997). In addition, strain-specific characteristics such as poor reproductive performance, developmental defects (e.g., colossal agenesis), infrequent use in the research community, and limited phenotypic strain characterization compared with C57BL/6 and BALB/c strains often limit the usefulness of 129 as a background strain. The creation of targeted mutant mice with C57BL/6J ES cells provides an attractive alternative. The existence of albino C57BL/6 mice (e.g., C57BL/6J-Tyr^c-2J/J; JAX® Mice stock number 000058) allows researchers to assess germ line transmission of the targeted mutation in chimeric mice and generate a virtually coisogenic strain within a single cross.

Temporal and tissue-specific regulation of gene expression enables researchers to analyze and/or circumvent phenotypes such as embryonic lethality and developmental defects, which are present in mice created by classical targeted mutagenesis techniques. Conditional and tissue-specific targeted mutations also can provide the proper controls for evaluating flanking DNA effects. Tetracycline-inducible systems (Baron and Bujard 2000; Furth et al. 1994) allow the study of phenotypes with genes in the “on” or “off” position. Similarly, cre-lox (Nagy 2000; Utomo et al. 1999) or FLP-FRT (Rodriquez et al. 2000) technologies permit phenotypic evaluation with and without ubiquitous or tissue-specific gene expression.

Breeding strategies to control for flanking DNA and genetic background effects have been proposed (Banbury Conference 1997; Wolfer et al. 2002). Using an elaborate breeding strategy to generate congenic strains and “reverse F1” and “reverse F2” hybrids, Chen and colleagues (2004) were able to isolate the effects of the elimination of the basigin transmembrane protein (Bsg) from genetic background and flanking gene effects.

Practical Considerations

Controlling the genetic background is not always practical or feasible. Many spontaneous mutations arise on a mixed or undefined genetic background. Historically, genetically diverse mouse stocks carrying mutations were inbred to create a novel strain. A classic example, V/Le, carries several recessive mutations including waltzer (Cad23^v), piebald (Ednrb^s), leaden (Mlph^ln), and fuzzy (fz). The obese mutation (Lep^ob) occurred in the V/Le strain (Ingalls et al. 1950) and was subsequently backcrossed to the inbred strains C57BL/6, C57BLKS, and BTBR. The obesity and diabetes phenotypes caused by Lep^ob vary dramatically on these three backgrounds (Coleman and Hummel 1973; Stoehr et al. 2000), indicating the presence of genetic modifiers.

Induced mutations (both targeted and random) and transgenics are often either generated or initially maintained on a mixed segregating or hybrid background (e.g., C57BL/6 and 129 for targeted mutations or C57BL/6 and SJL hybrid for transgenics). Further inbreeding or backcrossing to other inbred strains may result in a phenotype significantly different from the phenotype initially reported on the mixed genetic background. Interpreting such phenotypic differences is often difficult and requires the appropriate controls. Wild-type siblings provide the most appropriate controls for strains on mixed segregating backgrounds.

Thus, to return to Dr. Musmusculus' euphoria upon discovering that one of his knockout mice has a phenotype, we recall that he immediately recognizes the necessity of answering “difficult questions.” Despite a significant investment of time and skill (2 yr in gene targeting and chimeric mouse generation, and then another year to expand and age the knockout colony), he knows that still more work lies ahead before the genetic factors that influence his knockout mouse can be identified accurately. Only once that identification is complete will it be time to consider publishing the findings.

Summary

The creation of a genetically modified strain and its care, maintenance, and characterization necessitate the involvement of numerous personnel with differing expertise and knowledge. The use of proper nomenclature by all personnel will reduce time-consuming and costly ambiguities and errors. Genetically engineered mice may display phenotypes due either to specific mutations or to genetic background effects that affect their care and maintenance. Table 3 provides a summary of responsibilities and considerations related to genetic factors that affect phenotype. Use of proper nomenclature and an understanding of factors present in the genetic background of the mouse are critical for all personnel who work with genetically engineered mice, from the animal caretaker to the principal investigator.

Abbreviations used in this article: Chr, chromosome; ES, embryonic stem; NIH, National Institutes of Health; QTL, quantitative trait loci; tm, targeted mutation.

References

Accili D, Drago J, Lee EJ, Johnson MD, Cool MH, Salvatore P, Asico LD, Jose PA, Taylor SI, Westphal H. 1996. Early neonatal death in mice homozygous for a null allele of the insulin receptor gene. Nat Genet 12:106-109.

Alexander GM, Erwin KL, Byers N, Deitch JS, Augelli BJ, Blankenhorn EP, Heiman-Patterson TD. 2004. Effect of transgene copy number on survival in the G93A SOD1 transgenic mouse model of ALS. Brain Res Mol Brain Res 130:7-15.

Banbury Conference [Banbury Conference on Genetic Background in Mice]. 1997. Mutant mice and neuroscience: Recommendations concerning genetic background. Neuron 19:755-759.

Baron U, Bujard H. 2000. Tet repressor-based system for regulated gene expression in eukaryotic cells: Principles and advances. Methods Enzymol 327:401-421.

Barthold SW. 2002. “Muromics”: Genomics from the perspective of the laboratory mouse. Comp Med 52:206-223.

Beck JA, Lloyd S, Hafezparast M, Lennon-Pierce M, Eppig JT, Festing MF, Fisher EM. 2000. Genealogies of mouse inbred strains. Nat Genet 24:23-25.

Canning J, Takai Y, Tilly JL. 2003. Evidence for genetic modifiers of ovarian follicular endowment and development from studies of five inbred mouse strains. Endocrinology 144:9-12.

Chen S, Kadomatsu K, Kondo M, Toyama Y, Toshimori K, Ueno S, Miyake Y, Muramatsu T. 2004. Effects of flanking genes on the phenotypes of mice deficient in basigin/CD147. Biochem Biophys Res Commun 324:147-153.

Chevalier-Mariette C, Henry I, Montfort L, Capgras S, Forlani S, Muschler J, Nicolas JF. 2003. CpG content affects gene silencing in mice: Evidence from novel transgenes. Genome Biol 4:R53-1-R53-11.

Coleman DL, Hummel KP. 1973. The influence of genetic background on expression of the obese (Ob) gene in the mouse. Diabetologia 9:287-293.

Cormier RT, Bilger A, Lillich AJ, Halberg RB, Hong KH, Gould KA, Borenstein N, Lander ES, Dove WF. 2000. The Mom1 AKR intestinal tumor resistance region consists of Pla2g2a and a locus distal to D4Mit64. Oncogene 19:3182-3192.

Davisson MT. 2006. Genetic mapping. In: Fox J, Barthold S. Davisson MT, Newcomer C, Quimby F, Smith A, eds. Mouse in Biomedical Research. New York: Elsevier (In Press).

Erickson RP. 1996. Mouse models of human genetic disease: Which mouse is more like a man? Bioessays 18:993-998.

Festing MF, Simpson EM, Davisson MT, Mobraaten LE. 1999. Revised nomenclature for strain 129 mice. Mamm Genome 10:836.

Festing MFW. 2001. Inbred strains of mice. Mouse Genome Informatics, The Jackson Laboratory, Bar Harbor, Maine. World Wide Web (http://www.informatics.jax.org/), accessed March 2001 (http://www.informatics.jax.org/external/festing/search_form.cgi).

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.

Gurney ME. 1997. The use of transgenic mouse models of amyotrophic lateral sclerosis in preclinical drug studies. J Neurol Sci 152(Suppl 1):S67-S73.

Gurney ME, Pu H, Chiu AY, Dal Canto MC, Polchow CY, Alexander DD, Caliendo J, Hentati A, Kwon YW, Deng HX, Chen W, Zhai P, Sufit RL, Siddique T. 1994. Motor neuron degeneration in mice that express a human Cu,Zn superoxide dismutase mutation. Science 264:1772-1775.

Haines J, Johnson V, Pack K, Suraweera N, Slijepcevic P, Cabuy E, Coster M, Ilyas M, Wilding J, Sieber O, Bodmer W, Tomlinson I, Silver A. 2005. Genetic basis of variation in adenoma multiplicity in Apc^Min/+Mom1^S mice. Proc Natl Acad Sci U S A 102:2868-2873.

Ingalls KP, Dickie MM, Snell GD 1950. Obese, a new mutation in the house mouse J Hered 41:317-318.

International Committee on Standardized Genetic Nomenclature for Mice. 2005. Mouse Genome Informatics Web Site, The Jackson Laboratory, Bar Harbor, Maine. World Wide Web (http://www.informatics.jax.org/mgihome/nomen/strains.shtml#mis), accessed November 2005.

Jordan S, Beermann F. 2000. Nomenclature for identified pigmentation genes in the mouse. Pigment Cell Res 13:70-71.

Kido Y, Philippe N, Schaffer AA, Accili D. 2000. Genetic modifiers of the insulin resistance phenotype in mice. Diabetes 49:589-596.

Linder CC. 2002. Mouse nomenclature and maintenance of genetically engineered mice. Comp Med 53:119-125.

Maltais LJ, Blake JA, Chu T, Lutz CM, Eppig JT, Jackson I. 2002. Rules and guidelines for mouse gene, allele, and mutation nomenclature: A condensed version. Genomics 79:471-474.

Montagutelli, X. 2000. Determining the genetic basis of a new trait. In: Sundberg JP, Boggess D, eds. Systematic Approach to Evaluation of Mouse Mutations. Boca Raton: CRC Press. p 15-33.

Mouse Genome Database (MGD). 2005. Mouse Genome Informatics Web Site, The Jackson Laboratory, Bar Harbor, Maine. World Wide Web (http://www.informatics.jax.org/), accessed September 2005.

Nagy A. 2000. Cre recombinase: The universal reagent for genome tailoring. Genesis 26:99-109.

Nishina PM, Wang J, Toyofuku W, Kuypers FA, Ishida BY, Paigen B. 1993. Atherosclerosis and plasma and liver lipids in nine inbred strains of mice. Lipids 28:599-605.

Opsahl ML, McClenaghan M, Springbett A, Reid S, Lathe R, Colman A, Whitelaw CB. 2002. Multiple effects of genetic background on variegated transgene expression in mice. Genetics 160:1107-1112.

Padjen K, Ratnam S, Storb U. 2005. DNA methylation precedes chromatin modifications under the influence of the strain-specific modifier Ssm1. Mol Cell Biol 25:4782-4791.

Rodriguez CI, Buchholz F, Galloway J, Sequerra R, Kasper J, Ayala R, Stewart AF, Dymecki SM. 2000. High-efficiency deleter mice show that FLPe is an alternative to Cre-loxP. Nature Genet 25:139-140.

Silver LM. 1995a. General strategies for mapping mouse loci. In: Mouse Genetics: Concepts and Applications. New York: Oxford University Press. p 150-155.

Silver LM. 1995b. Town mouse, country mouse. In: Mouse Genetics: Concepts and Applications. New York: Oxford University Press. p 22-29.

Silverman KA, Koratkar R, Siracusa LD, Buchberg AM. 2002. Identification of the modifier of Min2 (Mom2) locus, a new mutation that influences Apc-induced intestinal neoplasia. Genome Res 12:88-97.

Simpson EM, Linder CC, Sargent EE, Davisson MT, Mobraaten LE, Sharp JJ. 1997. Genetic variation among 129 substrains and its importance for targeted mutagenesis in mice. Nat Genet 16:19-27.

Stoehr JP, Nadler ST, Schueler KL, Rabaglia ME, Yandell BS, Metz SA, and Attie AD. 2000. Genetic obesity unmasks nonlinear interactions between murine type 2 diabetes susceptibility loci. Diabetes 49:1946-1954.

Threadgill DW, Yee D, Matin A, Nadeau JH, Magnuson T. 1997. Genealogy of the 129 inbred strains: 129/SvJ is a contaminated inbred strain. Mamm Genome 8:390-393.

Tinkle BT, Jay G. 2002. Molecular biology, analysis, and enabling technologies: Analysis of transgene integration. In: Pinkert CA, ed. Transgenic Animal Technology: A Laboratory Handbook. New York: Academic Press. p 459-471.

Utomo AR, Nikitin AY, Lee WH. 1999. Temporal, spatial, and cell type-specific control of Cre-mediated DNA recombination in transgenic mice. Nat Biotechnol 17:1091-1096.

West DB, Boozer CN, Moody DL, Atkinson RL. 1992. Dietary obesity in nine inbred mouse strains. Am J Physiol 262:R1025-R1032.

Wolfer DP, Crusio WE, Lipp HP. 2002. Knockout mice: Simple solutions to the problems of genetic background and flanking genes. Trends Neurosci 25:336-340.