Emerging Diseases of Laboratory Rodents


Presented by Steven H. Weisbroth, DVM
Platform Presentation at the Tribranch, AALAS Annual Meeting
Adams Mark Hotel, Philadelphia, PA, June 3-5, 2002

Table of Contents

    Stage I. Domestication: 1880 – 1960
    Stage II. Gnotobiotic Derivation: 1960 – 1985
    Stage III. Eradication of the Indigenous Murine viruses: 1980 – 1996
    Stage IV. Post-Indigenous Disease: 1996 – The Present


First of all, I want to thank the organizers of this year's Scientific Program for the Tribranch Meeting for giving me this bully pulpit from which to discuss emerging diseases of laboratory rodents. I believe that my experience at a diagnostic laboratory has given me a unique perspective from which to observe the changing patterns of rodent and lagomorph diseases over the years. They really have changed and it is past time for us to recognize that they have. I would like to introduce a discussion of emerging diseases by first reviewing how the patterns of disease agent encounters have changed over the years- in order to appreciate where you are at present, it's necessary to know where you came from.

You are all familiar, of course, with the basic precepts of microbiologic assessment of the health status of laboratory animals. Development of this data on a repetitive schedule forms the objective basis on which to: 1) establish and/or reconfirm the ongoing microbiologic status of commercial and institutional rodent breeding colonies, 2) to develop institutional procurement standards for supplier eligibility based on animal health criteria, and 3) to continuously monitor the health status of institutional research animal resident colonies, including recent arrivals undergoing equilibration or quarantine prior to release for use, those currently involved in research protocols and those coming off study. The strategy of health surveillance is to detect by examination of one or more sample groups the presence of any pathogen from a specific profile of infectious agents. In an agent is detected in the sample group, even in a single individual, the inference can be made that the larger group, represented by the sample, is contaminated, as well. Of equal importance is the inability to detect any of a specific profile of agents under circumstances controlled for adequacy of the sample and detection technology. By this process the designated production or research unit may be demonstrated as free of the agents on the list.

Comprehensive rodent health surveillance programs are oriented to the systematic diagnostic examination of sample groups of animals against a predetermined list of pathogenic agents. In developing the lists of agents of concern, it is first necessary to understand that each of the laboratory species is host to an etiologic spectrum composed of arthropod ectoparasites, helminth and protozoan endoparasites, bacteria, viruses, rickettsial, and fungal forms typically associated by common diagnostic experience as indigenous to that species. For testing purposes the etiologic classes are organized into panels of the more common indigenous agents. In Table 1 I have summarized a comprehensive profile that, with minor exceptions, is widely used by rodent producers and users in the United States and which forms the basis for the FELASA panels used in Europe, as well. The profile forms the objective basis for a spectrum of agents that high quality research rodents are expected to be free of and is widely used as a procurement specification for that purpose.

Over the years, the principle effect of disease control, eradication, and exclusion programs has been to systematically reduce both the range of diversity and the frequency of encountering the agents of concern on this list. With the rodents, the following stages (summarized in Table 2) in the process can be recognized:

Stage I. Domestication: 1880 – 1960

As rodents were domesticated and brought into the laboratory, they brought with them the entire panoply of infectious agents associated with their wild counterparts. Under the improved conditions of sanitation, nutrition, interdiction of life cycles, and maintenance of breeding colonies within stable, indoor environments, a number of infectious conditions became progressively infrequent or simply disappeared form laboratory environments. Such agents as the rickettsias, numerous helminths, bacterial forms such as Leptospira, the rat bite fever agents Spirillum minus, and Streptobacillus moniliformis and even the Salmonella species serve as examples of agents that disappeared or which became notable by their infrequency under husbandry conditions that applied during the first half of the 20th Century.

Stage II. Gnotobiotic Derivation: 1960 – 1985

That is not to say that the general health of laboratory rodents or standards for the care of these animals in the 1950s and 1960s was all that good. Biomedical experimentation expanded greatly in the decades following World War II and with it a burgeoning if the utilization of laboratory rodents. The ravages of the primary mycoplasmal, bacterial, and viral pathogens imposed serious limitations on the successful conduct of research and testing programs reliant on the use of laboratory rodents. At a certain point in the 1960s it became clear that none of the standard veterinary medical approaches of improved husbandry and sanitation, vaccination, or antibiotic chemotherapy could effectively address the relentless effects of intercurrent disease on the research process. Surely, Mycoplasma pulmonis, the agent of respiratory mycoplasmosis was the agent that motivated the turn to a different approach to control or eradicate these diseases. The different approach was to prevent disease by excluding it. Henry Foster of Charles River Laboratories and C. N. Wentworth Cummings of Carworth Farms were among the first to perceive that the principles of gnotobiology, originated to explore the dimensions of germfree biology, could also be applied to large scale production of laboratory rodents from which the ineradicable diseases of the parents could be excluded by cesarean derivation.

The process is initiated by hysterectomy of the gravid uterus from the donor parent which is passed, by antiseptic immersion, into an isolator with sterile interior, life support systems and foster females with which to rear the neonates in the resected uterus. At some point the germfree neonates are associated with the 4-6 microbial forms necessary for normative intestinal physiology. These “associated” rodents, termed “gnotobiotes”, are retained in isolators as nucleus colonies used to produce progeny destined for transfer to large barrier production colonies. The offspring from within the barrier are offered for sale or for institutional use. In principle, such barriers may be continued in use indefinitely, unless routine testing indicates a “break” or penetration of the barrier by an unacceptable microbial form.

As the managers of commercial and institutional breeding colonies became adept with the principles and practice of gnotobiotic derivation, the process was adopted as the standard throughout the field and such terms as “pathogen free” and “specific pathogen free”, or “SPF&rduo; came into common parlance to describe the status of such animals. The process was remarkably effective as a means of producing rodents free of, at the time, common primary mycoplasmal, bacterial, and parasitic diseases, but less so with the murine viruses. At the time, little was known about the viral status of SPF animals, but more importantly, since there was little clinical perception of infection, there was correspondingly little concern about their presence in rodent hosts.

It was only after the aggregate burden of the primary parasitic, bacterial, and mycoplasmal agents had been substantially reduced that the more subtle morphologic and physiologic effects of the rodent viruses could be recognized. While, on the one hand, it is true that same derivation process that excludes parasitic, mycoplasmal, and bacterial agents also effectively excludes viruses, on the other, it is equally true that the Laboratory Animal Science establishment - the breeders, facility managers, and scientific users – in the decades from 1960 – 1980, knew about and tolerated viral enzootics out of indifference or uncertainty as to their health significance, or were simply unaware of their presence in production barriers and user facilities. In the absence of clinical signs and lesions, the effects of the murine viruses on their hosts could not be appreciated, or in most cases, even recognized. That state of affairs could not long last as a growing swell of research reports and conferences began defining the deleterious effects of the murine viruses. This concern ushers in the next stage, the unfinished business of the murine viruses.

Stage III. Eradication of the Indigenous Murine viruses: 1980 – 1996

The murine viruses, as a group, have only limited potential for serious clinical manifestation. In fact, most of the murine viruses were discovered or initially encountered as contaminants of research involving some transplantable neoplasm, tissue culture, or biologic derivative from infected, but clinically silent rodent hosts. Mouse pox (ectromelia) is practically the singular example of a virus with high potential for morbidity and mortality in immunocompetent mouse stocks. Several of the others, e.g. lymphocytic choriomeningitis virus (LCM), sialodacryoadenitis virus (SDAV), Sendai virus (SEN), mouse hepatitis virus (MHV), and Kilham's rat virus (KRV) serve as examples of agents that are ordinarily asymptomatic or with only moderate potential for pathogenicity in normative rodent stocks, but which have been indicted as agents of disease, in conventional terms, by some niche conferred by age group, genetic status or by rodents rendered inmmunodeficient by some heritable or transgenic process. Recent years have seen a great expansion of institutional holdings of immunodeficient, mutant, and transgenic stocks that have acted as flash points for clinical episodes in rodents having decreased resistance to agents that are ordinarily clinically silent in their normative rodent counterparts. Agents that include several of the Helicobacter species, the hyperkeratosis agent, Corynebacterium bovis, Pneumocystis carinii, and several of the murine viruses (MHV, MVM, PVM) are all extensively documented as clinically important disease hazards of this higher risk population. More important, by virtue of their clinical silence in immunocompetent rodent stocks are the many murine viruses that introduce some variability of cellular metabolism or reflexive cellular response to infection that interferes with the research process. Most of the murine viruses act as examples of this phenomenon.

By the early 1980s immense pressure from the research community, similar to that in the 1950s, began to develop as a result of too frequent viral complication of research in the areas of molecular biology and biotechnical research and product development. For about the last 15 years or so, an unofficial, but nonetheless consensually accepted international effort has been made to implement a virus free standard for the use of laboratory animals. It is important to understand that virtually every breeder and biomedical research institution in the country has had to come to grips with the scientific community's collective resolution to conduct research with virus free rodents. This effort has involved commercial and institutional rodent suppliers, the animal care unit administrators who procure and maintain rodent stocks, and the scientific users who conduct and report research results. In short, the entire chain that deals with the supply, care, and utilization of laboratory rodents.

Facilitating the process has been the wide availability of testing reagents to enable large scale health surveillance programs. This strategy, i.e., systematic and repetitive testing for the indirect serologic indicators of infection (antibodies) in sera from colony residents or sentinels has been universally adopted as standard practice to monitor viral contamination status of rodent colonies. The effort has been quite successful when viewed from a larger perspective. Over the last 10-15 years both the diversity of rodent viruses encountered and the frequency with which such infections are detected have declined markedly. We are, at present, in the terminal, mopping up stages of this international effort. It is also true that progress in viral eradication has not occurred without periodic shutdowns at user facilities and individually wrenching dislocations in research programs. Every animal care program manager has had to learn the techniques for prevention, control, and eradication of murine virus infections and how to deal constructively with the political balancing act required to achieve the support and compliance of institutional animal users. Most well managed animal care programs operate at present for long periods of time without serologic evidence of viral exposures and without detection of other indigenous agents by routine health surveillance. Laboratory animal specialists can take deserved pride in having guided the research animal effort in this country to the present level of quality. So, is the war over? Have the good guys won? Are the diagnostic labs to become like the Maytag repairman waiting for the phone to ring? Yes and no, but not really.

Stage IV. Post-Indigenous Disease: 1996 – The Present

Perversely, the diseases keep coming. What makes them different is that they either don't appear in Table 1 as amongst the agents defined by history and experience as indigenous to the particular rodent hosts, or they appear to biologically differ sufficiently as to suggest that something new and different is at hand. I am suggesting use of the term “post-indigenous” as a descriptor for an unfolding cluster of seemingly new conditions diagnostically taking shape as time goes on. They are summarized in Table 3.

Although admittedly based on circumstantial criteria, these conditions seem to have several common threads that have suggested themselves to me from my perspective at a diagnostic laboratory:

  1. Most of these conditions have only been recognized in the last several years. There is little, if any basis of prior experience or body of literature for understanding the provenance of these conditions. It is not clear if they are truly emergent, or rather, that they may now be more readily recognized in the absence of indigenous agents, or more likely, that both situations are at play. Surely the rodent helicobacters would fit this description.

  2. Given the recent arrival, or at least, recent recognition of these conditions, there is only a scanty base of literature to draw on for understanding the biology of these infections, which is the basis of strategies for their management and control. Certainly, this description would apply to RRV, in which even its recognition is hampered by lack of published pathologic descriptions to define the condition and its range of expression.

  3. The biology of infection with these conditions is not well understood, but anecdotal field reports, tracking data and eradication programs suggest that these conditions are not highly infectious, especially the many recent episodes of what are diagnosed as mouse MPV, MHV, and EDIM virus infections. In other words, as compared to their murine virus counterparts, these recent infections demonstrate reduced replicative ability to achieve animal-to-animal transfer, suggestive of a virus in an aberrant host. It is common experience that these infections are more easily contained and eradicated.

  4. On the whole, these infections are clinically mild or unapparent. Many have not been associated with lesions or physiologic changes, and several may only be recognized by seroconversion, seemingly without other concomitants or means of corroborating infection. All of the listed viral conditions would seem to fit this description. In fact, I'm reminded of an episode that Howard Blatt encountered some years ago at Albert Einstein College of Medicine (in NY) that he described as a nonpathogenic variant of MHV in athymic nude mice; i.e., the in-cage sentinels seroconverted but the nude mice remained healthy. I think that episode should now be properly classified as a nonindigenous, possibly human coronavirus infection stimulating antibodies in the sentinels read by diagnostic ELISA tests (using MHV antigens) as representing MHV infection.

  5. It is by no means clear for any of these conditions that the reservoirs of infection reside in other rodents- as is the case for all of the agents in the comprehensive panel listed in Table 1. That is to say, the reservoirs for typical rodent agents, e.g. Mycoplasma pulmonis, pinworms, mites, pasteurellas, and viruses such as Sendai are well established as residing in other rodents and almost always, some basis of contact with other rodents can be established in sorting out the history of the infection. This relationship is not clearly established with the cluster of emergent conditions being discussed here and indeed, a good deal of circumstantial and anecdotal evidence suggests otherwise.

If not from other rodents, where might such infections originate? This is an open question, at present, but I believe Occam's razor suggests the answer. Occam's razor states that when attempting the solution to a problem with several alternative explanations, the simplest explanation is usually correct. In my opinion, the source of most of the infections in this cluster is human and reflects the relatively open and unrestricted exposure of barrier-produced laboratory rodents to their human contacts; i.e. to the animal care personnel and investigators they come in contact with after arrival at the user institution. Of course, whether infection occurs depends on the chance concurrence of human virus shedder and proximity of susceptible rodent host, but the institutional carelessness typically exercised in rodent maintenance programs almost ensures a steady flow of virus infections- and that is what we see. As a whole, we have grossly underestimated and ignored the potential for communicability of human agents to laboratory animals. This casualness is one of the last unexamined and essentially unregulated aspects of the research animal environment. Surely it is an area we are going to have to look at more closely directly in the context of animal disease control. The rationale for doing so can be illustrated by the following points:

  1. Even under the best of circumstances to limit exposure, animals readily become colonized with microbial forms by virtue of human contact. Consider, if you will, a large commercial rodent production barrier with capability for sterilizing all incoming animal diets and beddings, cages, and related accessories, with HEPA filtration of incoming air, treatment of drinking water, and strict standards for personnel entry, including disrobing of street clothes, showering, donning of sterile outerwear consisting of jumpsuit, gloves, head cover, footwear, and surgical mask. When gnotobiotes are introduced into such an environment, it is commonly observed that is only a matter of 4-6-8 weeks before these animals become variably contaminated by a range of mammalian organisms that should be interdicted and prevented from entry by the barrier procedures. Such organisms include oral and enteric forms like E. coli, Aerobacter, Pseudomonas, and Klebsiella, and skin forms such as coryneforms, Staphylococcus epidermitis, and S. aureus. How could such bacterial forms gain entry to such an environment? Given the rigors of barrier SOPs, what are realistic possibilities? Under the circumstances described, the potential for a rodent source seems quite remote. Doesn't Occam's razor suggest the source as being human? It's reasonable to conclude that in spite of the rigorous use of barrier entry procedures, these organisms are being disseminated from animal care personnel, perhaps by microbe laden leakage of air from around the cuffs and masks. Or from small breaks in the rubber gloves (surely it is common experience that such breaks are sufficient to contaminate germfree isolators). Isn't it logical that if human-origin non-pathogens like E. coli can colonize barriered rodents, that pathogens like parvoviruses or rotaviruses could be vectored by the same human contacts as well?

  2. Another, more indirect example is suggested by experience at Taconic with a situation similar to the one above. In this case, there is a similarly well managed breeding barrier stocked initially with gnotobiotes, but here the caretaker personnel are provided with ventilated head cover equipment (e.g., Dryden covers) that fits closely over the head and neck like a diving helmet and which directs the expired air through a HEPA filter. The effect of this type equipment is to delay and, in some cases, prevent contamination of barrier rodent contacts with human agents like Klebsiella. This example underscores the comparative leakiness of standard outerwear and confirms suspicion that the likely sources of mammalian oral, enteric, and respiratory agents that contaminated the gnotobiotes in the first example are the animal care personnel having direct animal contact with them in the barrier.

  3. Another example of human reservoir is provided by common experience with closed guinea pig breeding colonies. Guinea pigs are prone to infection with human paramyxoviruses and are widely used in influenza research for this reason. Inadvertently, even barriered production colonies are prone to seroconvert to human origin paramyxoviruses and reoviruses (other than REO3) under circumstances where is no credible basis to posit exposure to other guinea pigs. The problem is that during routine health surveillance testing, the resultant sera, when tested by ELISA using extractives with broad virus family antigenic determinants, are serologically positive and read as representing antibodies to SV5, or Sendai and REO3, respectively. For this reason, some breeders use rat or mouse sentinels in guinea pig colonies to reflect the true status of the colony for the presence of SV5, Sendai, and REO3 viruses since rats and mice are less susceptible to human origin paramyxoviruses and reoviruses. The point is that the guinea pigs may be infected by these human agents, and if they are, the resultant antibodies may be uncritically interpreted as representing murine virus infections.

  4. Indirect evidence is further suggested by the infectivity patterns of certain viruses. In recent years at Taconic Anmed, serologic positives encountered in sera from a wide range of sources have mainly involved three agents: MHV, MPV, and EDIM. They have tended to occur in rodents variably following removal from a commercial or institutional breeding barrier and delivery to a user institution. That is not to say that breeders never have viral “breaks”, but it is to say that the occurrence is progressively unusual and with a steep epizootiologic gradient of infection following removal from the barrier. Why should that be so? I believe that the reason lies in the greater isolation from other rodents provided by barrier SOPs, but especially by the heightened usage of outerwear enforced by barrier entry protocol and the greatly constricted exposure to human traffic in breeding barriers. The message here should be clear, i.e., the potential for virus breaks, especially with human agents recognized by testing reagents as murine viruses, can be much reduced by a combination of more rigorous use of outerwear and imposition of limits to human traffic coming into contact with the animals.

There appears little doubt that, at present, Laboratory Animal Science is witnessing a major restructuring of the lists of indigenous pathogens of laboratory rodents. Contemporary commercial and institutional rodent supplier stocks have little, if any, realistic potential as reservoirs for these agents. The truly indigenous pathogens, those listed in the comprehensive profile in Table 1, have largely been eradicated, even at user institutions. The process has been driven by constant resupply from disease free production sources and the highly structured and microbially limited environments permitted by good laboratory animal management practice. Circumstances seem to suggest, with some exceptions, that the reservoirs for the cluster of agents listed here as “emergent” are probably humans coming in contact with the more microbially limited rodent hosts. If so, and if we mean to further limit and reduce “rodent” disease as we currently experience it, we are going to have to more carefully standardize and regulate such contacts more rigorously than we do at present.

Top of Page

Table 1


  1. Supplemental Diagnostic Screen: The full range of clinical examinations, gross necropsy, microbiologic isolations, and histopathology to isolate or otherwise detect the presence of the following microorganisms:

      (a) Arthropod ectoparasites. Genera include: Myobia, Myocoptes, Radfordia, Polyplax, Psorergates, Notoedres, Demodex, Liponyssus.
      (b) Helminth endoparasites. Genera include: Aspicularis, Syphacia, Hymenolepis, Trichosomoides.
      (c) Enteric protozoa. Genera include: Hexamita (Spironucleus), Giardia, Entamoeba, Trichomonads, Eimeria.
      (d) Bacteria:
      Group I: Primary Pathogens. Salmonella sp., Streptobacillus moniliformis, Corynebacterium kutscheri, Pasteurella pneumotropica, Streptococcus pneumoniae, HAC coryneform (athymics only).
      Group II: Secondary Opportunists. Pseudomonas aeruginosa, Staphylococcus aureus, Citrobacter freundii, Type 4280, Klebsiella oxytoca, Klebsiella pneumoniae, Bordetella bronchiseptica, Beta hemolytic Streptococcus.
      (e) Hemoprotozoans (Rickettsia). Hemobartonella, Eperythrozoon.
      (f) Mycoplasmas. Mycoplasma pulmonis, Mycoplasma arthritidis.
  2. Molecular Microbiology: A panel of tests conducted by PCR to detect the presence of certain agents including: Helicobacter sp., Pneumocystis carinii, Corynebacterium kutscheri, Clostridium piliforme.

  3. Virus Serology, Supplemental Panel: A battery of viral and mycoplasma agents whose presence is detected by various tests for antibodies in serum. The presence of the virus in the colony is indicated by positive (+) antibody findings.
Mouse Panel
(Test 201-129)
Virus Name (b) Rat Panel
(Test 201-109)
PVM Pneumonia Virus of Mice PVM
REO-3 Respiratory Enteric Orphan III REO-3
GD-7 Encephalomyelitis Group GD-7
SEN Sendai SEN
LCM Lymphocytic choriomeningitis LCM
MYCO Mycoplasma pulmonis MYCO
HAN Hantaan Virus HAN
  Sialodacryoadenitis Virus/Rat Coronavirus) SDAV/RCV
  Kilham's Rat Virus/Rat Parvovirus KRV/RPV
  Toolan's H-1 TH1
MVM Minute Virus of Mice  
MPV Mouse parvovirus  
MHV Mouse Hepatitis Virus (coronavirus)  
KV Kilham's Virus  
EDIM Epizootic Diarrhea of Infant Mice  
MAV Mouse Adenovirus  
ECTR Ectromelia  
POLY Polyoma  
MCMV Mouse Cytomegalovirus (Salivary Gland Virus)  
MTV Thymic Virus  
LDHV Lactic Dehydrogenase Elevating Virus  
CKUT Corynebacterium kutscheri CKUT
ECUN Encephalitozoon cuniculi ECUN
CPIL Clostridium piliforme CPIL
Top of Page

Table 2

Stages in Rodent Health Profiles

Stage Description Years Involved
I Domestication 1880-1960
II Gnotobiotic Derivation 1960-1985
III Eradication of the Indigenous Murine Viruses 1980-1996
IV Post-Indigenous Disease 1996-the Present

Table 3

Emerging Disease Conditions of Laboratory Rodents

  1. Helicobacter infections of mice, rats, and hamsters.
  2. Beta hemolytic Streptococcus infections of mice.
  3. Staphylococcus aureus infections of athymic (nude) mice.
  4. Corynebacterium bovis (Hyperkeratosis-associated coryneform or HAC) infections of athymic (nude) mice.
  5. Atypical parvovirus infections (MPV and RPV) of mice and rats.
  6. Atypical reovirus and paramyxovirus (parainfluenza) virus infections of guinea pigs.
  7. Atypical mouse hepatitis virus (MHV) infections of mice.
  8. Atypical mouse rotavirus (EDIM) infections of mice.
  9. Rat respiratory virus (RRV) infections of rats.
Top of Page