Online Issues

Journal Vol 49 (3)

Introduction

Microbes and the Evolution of Scientific Fancy Mice

Stephen W. Barthold

Stephen W. Barthold, DVM, PhD, is Director of the Center for Comparative Medicine at the University of California in Davis.

Address correspondence and reprint requests to Dr. Stephen W. Barthold, Director, Center for Comparative Medicine, University of California, Davis, One Shields Avenue, Davis, CA 95616 or email swbarthold@ucdavis.edu.

"Efforts to guard the mice against disease required stringent sanitation standards . . . "

Leonell C. Strong, "Inbred Mice in Science" ca. 1922, St. Stephen's College, New York

Most historical accounts involving the laboratory mouse focus on genetics, but it is clear that today's inbred mice (herein euphemistically referred to as "scientific fancy mice") have been significantly affected by microbial agents from their very beginning. The inbred laboratory mouse began its storied history 100 years ago, at the advent of the American scientific movement, but its foundations were built on literally thousands of years of purposeful breeding and trading of "fancy" mice.

A dominant character in the fancy mouse movement, and in the early foundations of laboratory mice, was the Japanese waltzing mouse (see historical reviews in Davisson and Linder 2004; Morse 1978; Rader 2004; Weller 1978; Yoshiki and Moriwaki 2006). Most accounts attribute the first inbred laboratory mouse (dilute brown nonagouti, a.k.a. dba and subsequently DBA) to Clarence Cook (C.C.) Little, a graduate student of William E. Castle at the Harvard Bussey Institute from 1907 to 1914. Castle was researching "experimental evolution" in mammals, from the perspective of coat color inheritance, using dogs, cats, guinea pigs, rabbits, and mice. He assigned mouse breeding to Little, who performed seminal studies on the genetic basis of coat color (Russell 1978).

While Little was chasing the genetic rainbows of coat color, others, particularly Leo Loeb and Ernest E. Tyzzer, were pursuing research on genetic susceptibility to transplantable carcinomas and sarcomas among "races" of Japanese waltzing mice (Loeb 1906, 1908; Tyzzer 1907a,b, 1909). Successful transplantation of tumors in these races of mice indicated that a significant degree of inbreeding had already occurred before purposeful genetic inbreeding. After completing his graduate studies, Little went to work in Tyzzer's laboratory at Harvard and began the task of "Mendelizing" the cancer problem, crossing his dilute brown mice with Tyzzer's tumor-susceptible Japanese waltzing mice (Little and Tyzzer 1916a,b).

After a brief stint with Tyzzer at Harvard, Little took his dilute brown mice to the Carnegie Institution of Washington at Cold Spring Harbor, New York. In the summer of 1919 Leonell Strong, a graduate student at Columbia under the Nobel geneticist Thomas Hunt Morgan, also went to Cold Spring Harbor, with the intent of using Little's inbred DBA mice for cancer genetic studies. However, no sooner had Strong arrived than an epizootic of "mouse paratyphoid" completely decimated Little's mouse colony (and Strong's research). Little was able to obtain a trio of DBA mice from Tyzzer, but they were quite old. Bereft of hope that the geriatric mice would reproduce, Little bequeathed them to Strong. Strong brought the mice back from oblivion, continued to inbreed them, and later created other inbred lines of mice, among which were the tumor-susceptible C3H strain, the long-lived CBA strain, and A strain mice. Because of the limited number of DBA mice, Strong embellished the gene pool for many of his mouse lines with wild mice captured from a nearby pigeon coop at Cold Spring Harbor (thus the agouti coloration of C3H mice, which were derived from wild mice crossed with albinos and dilute brown nonagoutis).

Upon graduating, Strong took a job as instructor at St. Stephen's (now Bard) College in New York. He brought his mouse lines with him, but they were nearly destroyed by a fire in his mouse house (a renovated chicken coop, retrofitted with a sheet metal floor to keep the rats out). The DBA line was once again nearly lost, but Strong managed to rebuild his mouse lines upon relocating to the Bussey Institute, and subsequently took them to the University of Michigan with C.C. Little (Strong 1978). In 1929 Little and Strong moved to Bar Harbor to establish the Roscoe B. Jackson Memorial Laboratory, which benefited from Strong's collection of mice and lines developed by Little, including C57 lines (derived from Abbie Lathrop's mouse stocks; more on her work below), and others. Thus, although C.C. Little is often credited with creating the first inbred mouse strain (DBA), if it were not for Tyzzer and Strong DBA mice would not exist today and infectious disease would have robbed Little of his fame. As it is, many mouse lines were no doubt lost along the way due to infectious disease susceptibility.

The exact genetic origins of the DBA mouse and of other inbred strains of mice have been lost to history, but it is clear that most are an amalgamation of genetic stock, much of which was provided by Abbie Lathrop, a retired schoolteacher in Granby, Massachusetts. Having failed at raising poultry, she turned to breeding mice (as well as rats, guinea pigs, and rabbits) for the "fancy" rodent trade that was flourishing at the time. She began her mouse breeding operation with only a few Japanese waltzing mice and expanded it to an "average daily census" of approximately 11,000 mice at the zenith of her mouse career. She acquired new genetic lines of mice from other fanciers as well as from periodic additions of wild mice. Inbreeding for coat color was not a foreign concept to her; her notes indicated that she had inbred lines up to the 12^th filial generation, and it is likely that she supplied a number of scientists with lines of mice that were at least partially inbred.

Lathrop noticed that some of her mouse lines had a propensity toward cancer, and she provided these lines of mice to cancer researchers, including Loeb, Tyzzer, Strong, and E.C. MacDowell. With her scientific background, Lathrop performed necropsies on her mice and kept detailed notes, including annotations on infectious diseases (tapeworms, diarrhea, and hepatic abscesses, among others). Lathrop's contributions to science were genuine, and she coauthored several publications, particularly with Leo Loeb (Morse 1978).

This historical narrative is relevant to this issue of the ILAR Journal as many of these genetic pioneers also studied and described—in some cases, for the first time—the impacts on their mice of various microbial agents. Ernest Tyzzer was a scientist whose involvement in cancer research was significant, but brief, as the Director of Research for the Harvard Cancer Commission. He then succeeded Theobald Smith as the Fabyan Professor and Head of the Department of Comparative Pathology at Harvard (Weller 1978). Like Smith, Tyzzer's primary interest was infectious disease, and some of his publications described, for the first time, Cryptosporidium of the stomach and intestines of mice (now C. muris and C. parvum, respectively) (Tyzzer 1910), and a new disease in his Japanese waltzing mice caused by a spore-forming intracellular bacterium, Bacillus piliformis (later to be renamed Clostridium piliforme) (Tyzzer 1917). The latter is now well recognized as "Tyzzer's disease" and has been documented in a wide variety of mammalian species (including laboratory rats, gerbils, hamsters, guinea pigs, and rabbits) (Percy and Barthold 2007). Tyzzer demonstrated that the agent was transmissible from soiled bedding (a concept practiced today with sentinel mice). Thus, Lathrop and Tyzzer recognized the association of infectious agents with their mouse colonies.

C.C. Little, too, learned the lessons of infectious disease when his DBA mouse colony at Cold Spring Harbor was devastated by "mouse paratyphoid," which, in retrospect, could very well have been Tyzzer's disease, since salmonellosis tends to be more insidious in mice (Percy and Barthold 2007). What is most remarkable about this outbreak was the uniform susceptibility of the entire inbred DBA population, a classic example of genetic susceptibility that is unique to inbred mouse populations. Leonell Strong also realized that his various lines of mice were often quite fragile due to inbreeding, and therefore advocated and practiced strict sanitation standards.

The cancer connection came full circle when Strong's graduate student John J. Bittner demonstrated an infectious "milk influence" (later identified as a mammary tumor virus) in mice that was responsible for early onset mammary cancer (Bittner 1936, 1942).

Others in these early years were also learning the lessons of infectious disease in their research. Jacob Furth, a pathologist at Columbia University, was among the first to realize that infectious agents interfere with research results in genetically inbred mice. He saw that his high-leukemia Ak (later to become AKR) mice experienced delays in tumor onset when they developed pneumonia and other infections (in retrospect due to stress-related lymphocytic apoptosis in the thymus, which is the target organ for the AKR retrovirus). Tyzzer's transplantable tumor line (JwA) originated as a salivary gland tumor in Japanese waltzing mice, and the tumor could very well have been the product of enzootic polyomavirus infection, which was not discovered until the 1960s but clearly was prevalent among laboratory mouse populations at that time. Considering the mouse husbandry conditions (barns, chicken coops, pigeon coops) maintained by Lathrop, Castle, Little, Tyzzer, Strong, and others, microbial pathogens were no doubt quite prevalent and there was ample opportunity for cross contamination among mouse stocks through both the trading of these "scientific fancy mice" and through purposeful or unintended introductions from feral and wild mice.

Awareness of infectious agents in laboratory rodents evolved with technology. D. Kutscher (1894) documented the first report of a bacterial disease in mice, "pseudotuberculosis" (caused by Corynebacterium kutscheri), which may have been the cause of the hepatic abscesses noted by Lathrop. Among the first agents to be recognized were the most clinically obvious (such as Cryptosporidium spp. by Tyzzer in 1912). Chronic respiratory disease (now ascribed to Mycoplasma pulmonis, cilia-associated respiratory bacillus, and other factors) was well recognized as a significant disease in rats as early as 1915 (Hektoen 1915-1916) and somewhat later in mice (Nelson 1937).

Researchers became aware of viruses in 1930, when a disease outbreak called infectious ectromelia caused amputation of extremities (dry gangrene) in surviving mice (Marchal 1930). Mouse encephalomyelitis virus was recognized in mice with "spontaneous" paralysis in 1934 (Theiler 1934), and the next year a veterinarian, Erich Traub of the Rockefeller Institution at Princeton, published very insightful studies on natural infection of mice with lymphocytic choriomeningitis virus (Traub 1935). In these early years, there was general awareness of epizootic diarrhea of infant mice (EDIM), but the rotavirus etiology was not yet known (Cheever and Mueller 1947a). During studies on EDIM, experimental mice developed posterior paresis, which led to the discovery in 1949 of JHM virus (Cheever et al. 1949), the harbinger of the very extensive MHV group of viruses (reviewed in Piazza 1969).

The 1950s and 1960s heralded the discovery of many more largely adventitious viruses—such as lactate dehydrogenase–elevating virus, cytomegalovirus, mouse adenovirus, minute virus of mice, polyomavirus, and reovirus, among others—that became apparent in studies involving cell culture, transplantable tumors, and cancer research. The contributions to rodent infectious disease knowledge of scientists such as Wallace Rowe, Janet Hartley, Sara Stewart, Bernice Eddy, Ludwig Gross, John B. Nelson, Lisbeth Kraft, and numerous others were significant during this expansive period of rodent history (reviewed by agent in Fox et al. 2007). These advances prompted the Centers for Disease Control and Prevention (CDC) and National Cancer Institute (NCI) to hold a symposium in 1965 on Viruses of Laboratory Rodents (Holdenried 1966). The state of awareness of rodent infectious agents at that time was rudimentary, as exemplified in the classic but out-of-print book Pathology of Laboratory Rats and Mice by Cotchin and Roe (1967). A subsequent landmark conference, Viral and Mycoplasmal Infections of Laboratory Rodents: Effects on Biomedical Research, was held at the National Institutes of Health (NIH) in Bethesda in 1984 (Bhatt et al. 1986). Advances in knowledge between these two conferences were significant, although now very much out of date. Additional, less extensive conferences have been held on the subject (NRC 2000). The rapid growth of infectious disease awareness among laboratory rodents prompted the publication in 1991 of an ILAR report on Infectious Diseases of Mice and Rats, which began to address the issue of infection versus disease, principles of rodent disease and prevention, and the effects of individual agents on research (Lindsey et al. 1991).

As research with the laboratory mouse has become increasingly sophisticated and technology has evolved, the list of known infectious agents that are not only overt but also opportunistic pathogens has grown and continues to grow. We now know that the laboratory mouse can be host to well over 60 different pathogens (closer to 100 if you count individual species) (Percy and Barthold 2007), and an even larger list of agents has been documented in wild mouse populations (M. domesticus and M. musculus) (Singleton and Krebs 2007). Much of this growth is from recognition of opportunistic pathogens in genetically altered mice, such as Helicobacter spp., but it is also due to advances in detection assays. For example, the discovery of an entirely new serogroup of mouse and rat parvoviruses accompanied the transition from hemagglutination inhibition serology to the immunofluorescence assay (IFA) and enzyme-linked immunosorbent assay (ELISA), which detected cross-reactive nonstructural antigens (Smith 2000). Molecular methods continue to expand diagnostic horizons, with the amplification of nucleic acids for detecting agents and the creation of recombinant antigens for diagnostic use.

The mouse poses unique challenges because of the nature of the beast: there are many inbred strains with intrinsic and extrinsic genetic susceptibility to various agents, mutant strains, and immunodeficient strains, and there is much more sophisticated research that is subject to relatively subtle experimental variables compared to other species. Thus, drawing the line between overt pathogens, opportunistic pathogens, and normal microflora/-fauna of the mouse can be daunting, if not impossible.

Where did all of these infectious agents come from, and why does the mouse seem to have so many? Consider the history of the laboratory mouse. It is a mosaic of Mus musculus subspecies, including M.m. domesticus, M.m. musculus, M.m. castaneus, M.m. molossinus (a natural hybrid of M.m. musculus and M.m. castaneus), and possibly others, depending on the strain (Yoshiki and Moriwaki 2006). The laboratory mouse mosaic genome is a consequence of the long history of global trading among mouse fanciers, journeys of commensal mice in concert with human global migration and colonization, and frequent introductions (intentional and not) of feral mice along the way (such as Strong's use of wild mice to create C3H mice and Lathrop's embellishments of her fancy mouse collection).

Laboratory mice have thus had the opportunity to acquire not only a staggering array of infectious agents that were indigenous to their diverse origins but also several new agents during domestication and trade among fanciers and scientists. Ectromelia virus, for example, first appeared in laboratory mice in England in 1930 and is now worldwide in distribution, but its natural origin remains an enigma. The periodic seasonal and multifocal outbreaks of Sendai virus in the 1970s suggested introductions to mouse populations through sources other than mice, such as human exposure. Debate continues about the origin (human or mouse) of Sendai virus, but recent studies suggest that it may indeed be of human origin (Skiadopoulos et al. 2002). Researchers have noted for years that rotaviral infection of mice (EDIM) has an unexplained seasonality (Cheever and Mueller 1947b).

The exchange of pathogens continues to plague mouse populations today through cross contamination during shipping, unrestricted traffic among institutions and investigators, exposure to untested biologic products, and continued introductions and reintroductions through uninformed investigators. There are also opportunities for the introduction or reintroduction of pathogens from aboriginal, commensal, and feral mice, as scientists seek to develop inbred lines of M.m. musculus, M.m. molossinus, M.m. castaneus, M. spretus, and other wild populations (Ike et al. 2007; Weller 1978).

Seldom considered is the profound influence of retroelements, ranging from evolutionarily recent replicating ecotropic retroviruses to ancient retrotransposons in the genome of the laboratory mouse. Collectively, retroelements make up over 37% of the total laboratory mouse genome (Mouse Genome Sequencing Consortium 2002) and contribute significantly to mouse strain predispositions to various types of cancer (for which, as indicated above, many strains of mice were originally selectively bred). Furthermore, retroviral integrations are responsible for several commonly recognized mouse strain phenotypes, including rodless retina (rd1, the recessive allele associated with retinal degeneration) in Strong's C3H mice, hairless mutation (hr), and the dilute (d) coloration of Little's (Lathrop's, Tyzzer's, and Strong's) DBA mice. Retroelement integrations contribute to nearly 15% of the spontaneous mutant phenotypes in laboratory mice, including athymia (Foxn1), stargazer (Cacng2), obese (Lep), and albino (Tyr), among many others. The mosaic genome of the laboratory mouse is reflected in retroelement mosaicism contributed by various M. musculus subspecies (reviewed in Percy and Barthold 2007). Take away retroviruses (and related ancient retroelements) and there is little left to call a mouse.

Rodent infectious disease was extensively controlled when most animals were obtained from commercial vendors in the 1970s and 1980s. For a brief period, rodents became remarkably pathogen-free thanks to competition among vendors in response to demands by the scientific community (Shek 2000). The "lists" of specified (known) pathogens were relatively short in "gnotobiotic" rodents during that era, and there were efforts to standardize rodent microflora. In the mid-1960s, Russell W. Schaedler created a "cocktail" of selected enteric bacteria from normal mice (Schaedler et al. 1965), and commercial breeders used these organisms to "associate" their rodent colonies with defined microflora (Baker 1966). Subsequently, the National Cancer Institute (NCI) modified the Schaedler flora ("altered Schaedler flora") in an attempt to standardize the microbial status of mice at all NCI contractors (Orcutt et al. 1987). These efforts to standardize the microflora of otherwise "pathogen-free" mice were laudable, but it is now clear that the population dynamics of mouse enteric flora, introduced in the form of eight species of standardized altered Schaedler flora, are significantly influenced by mouse genetic and environmental differences (Alexander et al. 2006). In other words, intestinal microflora can be standardized going in, but not going out.

With laboratory mice, there is no clear line of demarcation between "good" and "bad" when it comes to infectious agents. There is a tendency to condemn any virus as "bad," whether or not it is pathogenic or alters physiologic responses. Is norovirus "bad" if it never produces disease except in highly select circumstances, such as STAT1 deficiency? Is reovirus "bad" if, despite its ubiquity, it has a remarkably silent history of causing no natural disease and no effects on research (the only experimental evidence of reproducing reoviral disease was in conventional mice, which could not be reproduced in barrier-derived mice) (Barthold 1997)? How about parvoviruses? Mouse populations are often condemned on the basis of parvovirus seroconversion in a fraction of the population; but mouse parvovirus is so weakly pathogenic that even experimentally infected C57BL/6 mice often fail to seroconvert (Besselsen et al. 2000).

The challenge is even greater with bacteria. A great deal of effort and money are expended on control of Pseudomonas aeruginosa, but this ubiquitous environmental bacterium is pathogenic under very limited conditions (neutropenia induced by x-irradiation or cyclophosphamide). Alpha-hemolytic streptococcal bacteremia and mortality have been reported in irradiated immunodeficient mice, whose blood vessels fill with bacterial colonies but with virtually no host response (Percy and Barthold 2007). If we are concerned about Pseudomonas, why shouldn't we be concerned about excluding alpha-hemolytic streptococcus? Pasteurella pneumotropica is an opportunistic pathogen that is essentially normal intestinal microflora in wild and many conventional mice, but when it is eliminated its enteric niche is filled by Klebsiella, Proteus, or other enterobacteria. Proteus mirabilis can cause fatal disease in immunodeficient mice (Scott 1989); should it be excluded?

Should we test for Aspergillus terreus, Paecilomyces varioti, Pneumocystis murina, and other fungal agents? They are far more overt as pathogens in immunodeficient and genetically modified mice than many agents that are excluded. Tyzzer's Clostridium piliforme is often a target for testing, yet other pathogenic Clostridium spp. (such as C. difficile) are not. Reviews of mouse parasites invariably include Tritrichomonas spp. and Entamoeba muris, but there is no evidence that either has ever caused disease or adverse effects on research. On the other hand, Leptospira ballum, a significant zoonotic agent, is generally not considered, although it is very common in feral mouse populations, with ample opportunity for introduction into animal facilities.

Simply stated, laboratory animal science is having difficulty defining membership on the "list." In many cases, inclusion is historical and not rational, and in other cases agents that should be there are not, or should not be there but are. Harmonization is needed, with an assessment of what agents are relevant or significant based on scientific information.

Several years ago, Jon Gordon at Mount Sinai Medical Center in New York published a thought-provoking editorial in the journal Comparative Medicine about mouse infectious disease control programs, barrier maintenance, and their impediment to research. His premise was that laboratory animal programs were too restrictive for science and that they unnecessarily insisted on unrealistically maintaining pathogen-free rodents in agent-exclusive barriers (Gordon 2002). As he and others have pointed out (Barthold 2002, 2004; Franklin 2006; Morse 1986), the use of barrier conditions or rederived rodents may actually modify phenotypes and research results achieved under conventional conditions. Researchers have complained of this effect since long before the transgenic era. Immunologists have often noted that mice respond differently after rederivation, which results in a state of immunological hyporesponsiveness. Thus, mice can be "too clean," creating adverse effects on research. In today's genomic era, rederivation results in the phenomenon of "the disappearing phenotype," as occurred with genetically engineered mice with "inflammatory bowel disease" due to natural Helicobacter spp. infection (Barthold 2002).

Because most infectious "diseases" of laboratory rodents tend to be subclinical, there is naïveté or at least ambivalence in the scientific community about their importance, and so good surveillance and diagnostic programs, even if successful, are likely to be unappreciated. When funding gets tight and costs rise, the inevitable consequence is pressure to cut costs and eliminate quality control programs.

What role, then, does the laboratory animal professional community have in declaring what is "good" and "bad" for mouse research, beyond those agents that affect animal health and welfare? Are laboratory animal specialists the defenders of good science? Far too often, laboratory animal veterinarians find themselves in the role of policemen with very weak scientific justification to defend their position.

Why not open the gates and maintain research rodents in conventional facilities with good sanitation that prevents overt pathogens? Why not let mice acquire "wild-type" or "normal" microflora (including viruses)? The problem is not the concept but the mouse itself. The inherent value of the mouse is its "homogeneity of genetic constitution," as C.C. Little described it. Ignoring microbial variables essentially negates the value of the model, since it is irrefutable that adventitious infections induce significant effects on research in mice, and the inbred and mutant status of mice accentuates these effects. There is no such thing as "normal microflora" for the laboratory mouse because of its long and circuitous history of domestication. Laboratory mice are not only genetic mosaics but also microbial mosaics, so that there is no such thing as a "wild-type" mouse. Even if we chose to purposely reintroduce microflora in mouse populations, the task of ensuring their continued and uniform presence would be daunting, as has been shown with the mere eight bacteria that make up altered Schaedler flora. Unfortunately, exclusion is the most practical solution, which requires surveillance, diagnosis, control, and management.

The advent of transgenesis, with the development of many mutants with considerable value to specific areas of scientific inquiry but little global commercial value, has given rise to a new era of the "scientific fancy mouse." As the United States embarks on its NIH-funded Knockout Mouse Project and the NIEHS Environmental Genome Project (Comparative Mouse Genomics Centers Consortium), Europe on its European Conditional Mouse Mutagenesis Programme (EUCOMM), Canada its North American Conditional Mouse Mutagenesis Programme (NorCOMM), Japan doing likewise, and China rapidly entering the fray, the quality control and traffic of scientific fancy mice will be challenging. It is not likely and indeed unrealistic to expect that international traffic of the anticipated tens of thousands of new mouse lines, germplasm, and embryonic stem (ES) cells can be harmonized in terms of excluding pathogens. Even with concerted international agreement, standardization and quality control of testing modalities among countries and institutions would be unattainable.

Although it may seem that microbial surveillance and control is a Gordian knot with no solutions, it is really not that complicated. First and foremost is the rule of "buyer beware," which has always been the case with laboratory rodents. "Specific pathogen free" is a notoriously nebulous phrase, and meaningless without specifying the pathogens. Furthermore, the credibility of test results is vital, but these are affected by a variety of factors such as the testing laboratory, test method, sample size, animal age, and sentinel or indigenous animals. It is important to interpret results in context, not only who created them and how but also their ecology in the mouse population. For example, a single MHV seropositive sample among many negatives is suspect, whereas not so with less contagious agents like mouse encephalomyelitis virus or parvovirus. Mice that are clinically ill due to MHV infection are likely to be seronegative, whereas seroconversion is a confirmatory result for mice that are clinically ill from Sendai virus. Serology is meaningless for enzootic LCMV infection, as mice are tolerant to the virus and do not seroconvert. All too often, however, programs rely simply on serology and fail to recognize the value of diagnostic pathology to detect infections before they have spread throughout a population and resulted in seroconversion, or to discover (groan) new pathogens.

Rodent infectious disease surveillance and control require an understanding of the biology of each agent in context. Such surveillance is a specialty unto itself that tends to be underemphasized or inadequately addressed by laboratory animal programs. Unfortunately, there is a national shortage of qualified laboratory animal veterinarians to fulfill this need, as documented in a recent ILAR report (NRC 2004a).

Another essential element of infectious disease management is education. Infectious disease surveillance and control affect scientists and their research. Because scientists pay the bills they are entitled to information about rodent infectious agents, how they may alter research results, and why animals need to be quarantined, tested, and managed for the scientist's own protection as well as his or her peers. Informed scientists become advocates, whereas forcing restrictions and limiting access to their animals garners resentment.

A single standard of infectious agent exclusion seldom works for academic institutions, where traffic is difficult to control and animals may be maintained in conventional, barrier, or modified barrier conditions. There is little point, for example, in depopulating a room of mice with MHV if it is a conventional animal room with unrestricted traffic, yet all too often investigators are forced to shut down their research to eliminate MHV, only to experience its rapid reintroduction after weeks or months of inconvenience. Unless there is a plan to prevent reintroduction, including the education of scientists and their personnel, MHV will be back. This scenario breeds false expectations, resentment, and outright hostility in the absence of the education of the scientist, who must be enfranchised in the process.

There is no debate that infectious agents affect research in a variety of ways, and a central element of research is reproducibility. Thus, journals need to be encouraged to compel authors to include the "specifics" of the pathogen status of their research animals as well as the genetic background. Authors are accustomed to including details of various methods and reagents, down to specific genes and sequences, but frequently ignore or omit details of animal genetic background and environment. At best, microbial status is often described simply as "SPF." Journal editors continue to be reticent about dealing with these issues.

The rising specter of international traffic of mice, germplasm, and ES cells poses both a challenge and an opportunity. The challenge will be dealing with the diversity and varied quality of material that is created by and exchanged among "cottage industry" research laboratories rather than commercial vendors. In response to this challenge, the NIH National Center for Research Resources (NCRR) established a network of regional Mutant Mouse Regional Resource Centers (MMRRCs; http://www.mmrrc.org/) that serve the international biomedical research community as a repository of genetically engineered mice, germplasm, and ES cell lines with value to biomedical research but with low commercial value. The MMRRCs are regionally located at the University of California at Davis, the University of Missouri, and the University of North Carolina, and are centrally linked through an Informatics Coordinating Center. The MMRRC network provides a number of ancillary services to the scientific community such as cryopreservation, rescue of failing or aged lines through assisted breeding methods (e.g., intracytoplasmic sperm injection and in vitro fertilization), speed congenics, genotyping, construct design, electroporation, microinjection, health screening, pathology, phenotyping, colony management, and disease consultation. The repositories accept, cryopreserve, resuscitate, rederive, and distribute mouse lines to requesting investigators throughout the world.

The MMRRCs also offer the opportunity to provide genetic and microbial quality control of diverse mouse populations and materials during the process of acquisition and distribution, thereby relieving recipient institutions of the burden of quarantining and rederiving mice from noncommercial sources. The MMRRC network is linked internationally through the Federation of International Mouse Resources (FIMRe), which, in addition to the MMRRC, includes the Jackson Laboratory, the NCI Mouse Models of Human Cancers Consortium, the Canadian Mouse Consortium, the Canadian Mouse Mutant Repository, the European Mouse Mutant Archive, RIKEN BioResource Center in Japan, the Center for Animal Resources and Development (Kumamoto University), the Australian Phenomics Facility, and new members continually joining the Federation.

It is clear that the scientific environment is rapidly becoming global, with significant impact on biomedical research. There is a need for science-based international harmonization of a variety of laboratory animal health and welfare issues, and of the transportation and exchange of material involving all laboratory animal species. ILAR convened the International Workshop on the Development of Science-based Guidelines for Laboratory Animal Care in November 2003 in Washington, DC (NRC 2004b). This conference set the stage for further and more focused discussions in an upcoming meeting to be held at the National Academies in Washington on September 23-26, 2008, on Animal Research in a Global Environment: Meeting the Challenges. In addition, ILAR is embarking on updating the Guide for the Care and Use of Laboratory Animals, which will be challenged with the task of dealing with infectious disease quality control in laboratory rodent facilities. However, the last edition of the Guide (NRC 1996, 60) remains timely and appropriately summarizes how to deal with microbial agents in rodents:

Scientific objectives of a particular protocol, the consequences of infection within a specific strain of rodent, and the adverse effects that infectious agents might have on other protocols in a facility should determine the characteristics of rodent health-surveillance programs and strategies for keeping rodents free of specific pathogens.

Acknowledgment

Supported by grant number U42 RR14905, "University of California Davis Mutant Mouse Regional Resource Center," from the National Center for Research Resources (NCRR), a component of the National Institutes of Health (NIH).

References

Alexander AD, Orcutt RP, Henry JC, Baker J Jr, Bissahoyo AC, Threadgill DW. 2006. Quantitative PCR assays for mouse enteric flora reveal strain-dependent differences in composition that are influenced by the microenvironment. Mamm Genome 17:1093-1104.

Baker DE. 1966. The commercial production of mice with a specified flora. Natl Cancer Inst Monogr 20:161-166.

Barthold SW. 1997. Reovirus type 3 infection, liver, mouse. In: Jones TC, Popp JA, Mohr U, eds. Monographs on Pathology of Laboratory Animals: Digestive System. Berlin: Springer-Verlag. p 196-200.

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

Barthold SW. 2004. Intercurrent infections in genetically engineered mice. In: Holland EC, ed. Mouse Models of Human Cancer. Hoboken: Wiley. p 31-41.

Besselsen DG, Wagner AM, Loganbill JK. 2000. Effect of mouse strain and age on detection of mouse parvovirus 1 by use of serologic testing and polymerase chain reaction assays. Comp Med 50:498-502.

Bhatt PN, Jacoby RO, Morse HC III, New AE. 1986. Viral and Mycoplasmal Infections of Laboratory Rodents: Effects on Biomedical Research. Orlando: Academic Press.

Bittner JJ. 1936. Some possible effects of nursing on the mammary gland tumor incidence in mice. Science 84:162.

Bittner JJ. 1942. Milk-influence of breast tumors in mice. Science 95:462-463.

Cheever FS, Daniels JB, Pappenheimer AM, Bailey OT. 1949. A murine virus (JHM) causing disseminated encephalomyelitis with extensive destruction of myelin: I. Isolation and biologic properties. J Exp Med 90:181-194.

Cheever FS, Mueller JH. 1947a. Epidemic diarrheal disease of suckling mice. I. Manifestations, epidemiology, and attempts to transmit the disease. J Exp Med 85:405-416.

Cheever FS, Mueller JH. 1947b. Epidemic diarrheal disease of suckling mice. III. Effect of strain, litter, and season upon the incidence of disease. J Exp Med 88:309-316.

Cotchin E, Roe FJC. 1967. Pathology of Laboratory Rats and Mice. Oxford: Blackwell Scientific Publications.

Davisson MT, Linder CC. 2004. Historical functions. In: Hedrich HJ, Bullock G, Petrusz P, eds. The Laboratory Mouse (Handbook of Experimental Animals). London: Elsevier. p 15-24.

Fox JG, Barthold SW, Davisson MT, Newcomer CE, Quimby FW, Smith AL. 2007. The Mouse in Biomedical Research: Diseases, 2^nd ed, vol II. San Diego: Academic Press/Elsevier.

Franklin CL. 2006. Microbial considerations in genetically engineered mouse research. ILAR J 47:141-155.

Gordon JW. 2002. Barrier facilities for transgenic rodents in academic centers: A two-edged sword. Comp Med 52:397-402.

Hektoen L. 1915-1916. Observations on pulmonary infections in rats. Trans Chicago Pathol Soc 10:105-109.

Holdenried R, ed. 1966. Viruses of Laboratory Rodents, vol 20. Washington: US Government Printing Office.

Ike F, Bourgade F, Ohsawa K, Sato H, Morikawa S, Saijo M, Kurane I, Takimoto K, Yamad YK, Jaubert J, Berard M, Nakata H, Hiraiwa N, Mekada K, Takakura A, Itoh T, Obata Y, Yoshiki A, Mantagutelli X. 2007. Lymphocytic choriomeningitis infection undetected by dirty-bedding sentinel monitoring and revealed after embryo transfer of an inbred strain derived from wild mice. Comp Med 57:272-281.

Kutscher D. 1894. Ein Beitrag zu Kenntniss der bacillaren Pseudotuberculose der Nagethiere. Z Hyg Infektkr 18:327-342.

Lindsey JR, Boorman GA, Collins MJ Jr, Hsu C-K, VanHoosier GL Jr, Wagner JE. 1991. Infectious Diseases of Mice and Rats. Institute of Laboratory Animal Resources, National Research Council. Washington: National Academy Press.

Little CC, Tyzzer EE. 1916a. Further experimental studies on the inheritance of susceptibility to a transplantable tumor, carcinoma (J.w.A.) of the Japanese waltzing mouse. J Med Res 33:393-453.

Little CC, Tyzzer EE. 1916b. Studies on the inheritance of susceptibility to a transplantable sarcoma (J.w.B.) of the Japanese waltzing mouse. J Cancer Res 1:387-389.

Loeb L. 1906. Development of a sarcome and carcinoma after the inoculation of a carcinomatous tumor of the salivary gland of Japanese mouse. University of Pennsylvania Medical School Bulletin 19:113-116.

Loeb L. 1908. Uber Enstanhug sines Sarcoms nachs Transplantation eines Adenocarcinoms einer japanischen Maus. Zeitschrift auf Krebforschung 7:80.

Marchal J. 1930. Infectious ectromelia: A hitherto undescribed virus disease of mice. J Pathol Bacteriol 33:713-718.

Morse HC III, ed. 1978. Origins of Inbred Mice: Proceedings of a Workshop. Orlando: Academic Press.

Morse HC III. 1986. Unexpected biologic consequences of rederivation of mouse strains. In: Bhatt PN, Jacoby RO, Morse HC III, New AE, eds. Viral and Mycoplasmal Infections of Laboratory Rodents: Effects on Biomedical Research. Orlando: Academic Press. p 689-692.

Mouse Genome Sequencing Consortium. 2002. Initial sequencing and comparative analysis of the mouse genome. Nature 420:520-562.

NRC (National Research Council). 1996. Guide for the Care and Use of Laboratory Animals. Washington: National Academy Press.

NRC. 2000. Microbial Status and Genetic Evaluation of Mice and Rats. Washington: National Academy Press.

NRC. 2004a. National Need and Priorities for Veterinarians in Biomedical Research. Washington: National Academies Press.

NRC. 2004b. The Development of Science-Based Guidelines for Laboratory Animal Care: Proceedings of the November 2003 International Workshop. Washington: National Academies Press.

Nelson JB. 1937. Infectious cattarh of mice. I. A natural outbreak of the disease. J Exp Med 65:833-842.

Orcutt RP, Gianni FJ, Judge RJ. 1987. Development of an "altered Schaedler flora" for NCI gnotobiotic rodents. Microecol Ther 17:59.

Percy DH, Barthold SW. 2007. Pathology of Laboratory Rodents and Rabbits, 3^rd ed. Ames: Blackwell Publishing.

Piazza M. 1969. Experimental Viral Hepatitis. Springfield IL: Charles C. Thomas.

Rader K. 2004. Making Mice: Standardizing Animals for American Biomedical Research, 1900-1955. Princeton: Princeton University Press.

Russell ES. 1978. Origins of mouse inbred strains: Contributions of Clarence Cook Little. In: Morse HC III, ed. Origins of Laboratory Mice. New York: Academic Press.

Schaedler RW, Dubos R, Costello R. 1965. Association of germfree mice with bacteria isolated from normal mice. J Exp Med 122:77-82.

Scott RAW. 1989. Fatal Proteus mirabilis infection in a colony of SCID/bg immunodeficient mice. Lab Anim Sci 39:470-471.

Shek W. 2000. Standardization of rodent health surveillance: Regulation versus competition. In: Microbial Status and Genetic Evaluation of Mice and Rats. Washington: National Academy Press. p 11-15.

Singleton GR, Krebs CJ. 2007. The secret world of wild mice. In: Fox JG, Barthold SW, Davisson MT, Newcomer CE, Quimby FW, Smith AL, eds. The Mouse in Biomedical Research. I. History, Wild Mice, and Genetics, vol 1. San Diego: Academic Press/Elsevier. p 25-51.

Skiadopoulos MH, Surman SR, Riggs JM, Elkins WR, St Claire M, Nishio M, Garcin D, Kolakofsky D, Collins PL, Murphy BR. 2002. Sendai virus, a murine parainfluenza virus type 1, replicates to a level similar to human PIV1 in the upper and lower respiratory tract of African green monkeys and chimpanzees. Virology 297:153-160.

Smith AL. 2000. Emerging (and reemerging) viruses of laboratory mice and rats. In: Microbial Status and Genetic Evaluation of Mice and Rats. Washington: National Academy Press. p 27-34.

Strong LC. 1978. Inbred mice in science. In: Morse HC III, ed. Origins of Inbred Mice. New York: Academic Press.

Theiler M. 1934. Spontaneous encephalomyelitis of mice: New virus disease. Science 80:122.

Traub E. 1935. A filterable virus recovered from white mice. Science 81:298-299.

Tyzzer EE. 1907a. The inoculable tumors of mice. J Med Res 18:137-153.

Tyzzer EE. 1907b. A study of heredity in relation to development of tumors in mice. J Med Res 17:199-207.

Tyzzer EE. 1909. A study of inheritance in mice with reference to their susceptibility to transplantable tumors. J Med Res 21:193-247.

Tyzzer EE. 1910. An extracellular coccidium, Cryptosporidium muris (gen. et sp.nov.) of the gastric glands of the common mouse. J Med Res 23:487-509.

Tyzzer EE. 1917. A fatal disease of the Japanese waltzing mouse caused by a spore-bearing bacillus (Bacillus piliformis, n.sp.). J Med Res 37:307-338.

Weller TH. 1978. Ernest Edward Tyzzer, 1875-1965. Washington: National Academy of Sciences.

Yoshiki A, Moriwaki K. 2006. Mouse pheromone research: Implications of genetic background. ILAR J 47:94-102.