Microbial and Phenotypic Definition of Rats and Mice: Proceedings of the 1998 US/Japan Conference (1999)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Development of Rodent Pathogen Profiles and Adequacy of Detection Technology Steven H. Weisbroth President, AnMed/Biosafe, Inc. Rockville, Maryland Dr. Ralph Dell, attendees of the US/Japan Meeting, and readers, as well, will have interest in being reminded that Dr. Howard Schneider, then Chairman of ILAR, in his opening remarks to an international symposium said that “Those familiar with the scientific process will recognize at once that this is only tempo- rary, that one can confidently predict further progress by the time of the Fifth International Symposium.” The symposium he was addressing was the Fourth, subtitled “Defining the Laboratory Animal,” given here in this city in 1969, and eventuating in a text of the proceedings under the same name (ICLA 1971). Thus, 30 years later, here we are discussing the same themes under the same subtitle. However, since then, a great deal of progress has been made, as pre- dicted by Dr. Schneider. I paraphrase my colleague Dr. David Baker who, in a recent review (Baker 1998), characterized progress in control of infectious dis- ease in laboratory rodents as follows: Around the turn of the century, an investigator might have said: “I can’t do my experiment today, all of my rats are dead.” In the 1960’s he might have said: “I can’t do my experiment today, all of my rats are sick”; and in the 1990’s he might have said: “I can’t do my experiment today, all of my rats are antibody positive.” A cynic might predict that in the year 2000 an investigator may have to say, “I can’t do my experiment today, my IACUC won’t let me use rats.” (This last statement, of course, is not attributable to Dr. Baker.) From the beginning, more than 100 years ago, until the present, issues of rodent health have remained as important concerns, not only to investigators, but 28 

STEVEN H. WEISBROTH 29 also to the service personnel charged with the production, procurement, and care of these animals. In attempting to put dimensions on the diversity of laboratory animal disease, it is first necessary to understand that each animal species is host to an etiologic spectrum composed of arthropod ectoparasites; helminth and proto- zoan endoparasites; and fungal, bacterial, rickettsial, and viral forms more or less associated by common experience and as documented in the literature as indig- enous to that host species. What the history of laboratory rodents seems to demonstrate is that this spectrum as a concept is not a list frozen for all time, but rather more closely represents a moving boundary in which old pathogens are eradicated, creating invasive opportunities for new pathogens and, thus, periodic reconstitution of the lists (Weisbroth 1996). In practice, the process moves gradually and the cast of characters is adjusted as circumstances and diagnostic experience warrant. Over the years, the principal effects of disease control, eradication, and exclusion programs have been to reduce both the range of diver- sity and the incidence of agents listed in the panels. There appears little doubt that at present we are witness to a major restructuring of the list of indigenous pathogens of laboratory rodents, a process accelerated by the highly structured and microbially limited environments permitted by production environments and good laboratory animal practice. For the purpose of defining the microbial health status of laboratory rodents and lagomorphs, comprehensive health surveillance programs are oriented to the systematic diagnostic examination of sample groups of animals against a pre- determined list of pathogenic organisms. The pathogens are organized into etiologic classes to form panels of the more common (or classically associated) indigenous agents. Collectively, these panels form the microbial definition of the status high quality research rodents are expected to meet. Sample groups statis- tically representative of the larger group from which they are drawn and meant to define are tested for the presence, or absence, of the specific agents making up the lists or panels. Properly conducted, findings in the sample groups can be used to infer presence of the detected agents in the larger population they represent. For purposes of monitoring the status of closed breeding colonies and of resident populations at user institutions, a program oriented to scheduled repeti- tive testing on an ongoing basis is developed. It is difficult to overstate the importance of scheduled, repetitive testing that not only provides current, timely information about the health status of specific subpopulations, but also because results on the sample groups form an additive sample size over time, increasing confidence that negative results in the sample groups truly are representative of the population as a whole. This concept is particularly important in assessing the reliability or accuracy of negative results. The strategy of health surveillance testing is oriented to detection of even a single positive instance, since such a finding implies that the larger population has been likewise exposed and may be contaminated with that agent. Conversely, failure to detect, or negative results for an agent, form the objective basis on which to conclude that the unit has not 

30 MICROBIAL AND PHENOTYPIC DEFINITION OF RATS AND MICE been exposed to that agent. Scheduled repetitive testing increases the sample size drawn from the unit and strengthens confidence that continued negative results are not a product of sampling error or test inadequacy. It is also important to recognize that the agents vary in prevalence; thus for closed units, it is unnecessary and uneconomical to uncritically and evenly apply the same testing frequency to all agents in the maximal panels. These consider- ations, so-called “smart testing,” lead health monitoring program managers to use reduced or “core” panels to monitor the more prevalent agents on a more frequent basis and maximal comprehensive panels on a less frequent basis, depending on the needs of the particular program. I would like to turn our attention now to a consideration of the agents making up testing panels themselves and adequacy of testing methodology to detect these agents, at this point in time. I hasten to explain that while the panels presented here have no official standing, they would with minor exceptions represent a consensus of expert opinion in the United States. The genera of important rodent louse and mite ectoparasites are listed in Table 1. Direct low power microscopic examination (10×) of cadavers shortly after euthanasia, at the level of the base of the hairs and skin, has proven to be the most rapid and accurate means of assessing ectoparasitic status. A more time- consuming, but technically acceptable, alternative is to allow the cadavers to cool while on black paper, allowing motile forms to come out to the surface or crawl off onto the paper where they may be picked up by cellophane tape, placed on slides, and identified under the microscope. Examination of skin scrapings is not TABLE 1 Arthropod Ectoparasites—Diagnostic Alternatives for Rodent Comprehensive Health Surveillance Profiles Acceptable Alternate Not Ectoparasite Standard or Adjunctive Recommended Myobia 1 2 3 Myocoptes 1 2 3 Radfordia 1 2 3 Psorgates 1, 3 Notoedres 1 2 3 Demodex 3 4 Liponyssus 1 2 3 Polyplax 1 2 3 1. Direct visualization of skin and pelage by low power microscopy 2. Motile forms on hair tips and on black paper under cooling carcass 3. Skin scraping 4. Skin section 

STEVEN H. WEISBROTH 31 a reliable method to establish ectoparasitic status. In general, for the arthropod parasites, present test methodology is adequate for surveillance programs. Table 2 is a listing of helminth endoparasites. The pinworms Syphacia and Aspicularis may be diagnosed alternatively by direct examination of the dis- sected cecum and colon by low power (3 to 30×) microscopy, or by fecal flotation, or both. Fecal flotation has the advantage of allowing simultaneous detection of (Eimeria) coccidia but could miss an early, preovulatory helminth infestation. As with the ectoparasites, present methodology is satisfactory for surveillance programs. The protozoa of importance are listed in Table 3. Readers familiar with the Federation of European Laboratory Animal Associations (FELASA) agent panels (Kraft and others 1994; Rehbinder and others 1996) may note the absence of Klossiella from this list, which has been deleted because it has not been reported from US colonies in modern times. There will be more to say about the FELASA listings as I continue. With the exception of the coccidia, the other enteric protozoa on this list are motile flagellates easily detected in temporary wet mounts of intestinal scrapings by microscopic examination (100×) or in histologic sections of small intestine. The coccidia, however, require fecal flotation for accurate detection of low-level infection. The hemoprotozoa require blood films for detection of parasitized cells but may be inapparent in latently infected immuno- competent hosts. The hemoprotozoa have not been reported in this country in many years, like lactic dehydrogenase elevating virus (LDHV), probably because of essential eradication of their (required) hematophagous arthropod vectors. Encephalitozoon infection is easily detected by present enzyme-linked immunosorbent assay assay (ELISA) serology. The latter test is sufficiently reliable to form the basis for “test and removal” eradication programs. Histo- pathology for Encephalitozoon detection is a useful confirmatory adjunct, but TABLE 2 Helminth Endoparasites—Diagnostic Alternatives for Rodent Comprehensive Health Surveillance Profiles Acceptable Alternate Not Helminth Standard or Adjunctive Recommended Aspicularis 1 3 4 Syphacia 1 3 4 Hymenolepis 3 4 Trichosomoides 2 4 1. Direct visualization of lumen contents of cecum and colon by low power microscopy 2. Direct visualization of lumen surface of urocyst by low power microscopy 3. Fecal flotation in hypertonic solution 4. Microscopic examination of histologic sections of relevant tissues 

32 MICROBIAL AND PHENOTYPIC DEFINITION OF RATS AND MICE TABLE 3 Protozoa—Diagnostic Alternatives for Rodent Comprehensive Health Surveillance Profiles Acceptable Alternate Not Protozoa Standard or Adjunctive Recommended A. Enteric Forms Giardia 1 3 Spironucleus (Hexamita) 1 3 Entamoeba 1 3 Trichomonas, Tritrichomona 1 3 Eimeria 4 1, 3 B. Hemoprotozoa Hemobartonella 2 Eperythrozoon 2 C. Other Encephalitozoon 5 3 1. Microscopic examination of wet mounts of intestinal scrapings 2. Microscopic examination of stained blood films 3. Microscopic examination of histologic sections of relevant tissues 4. Fecal flotation in hypertonic solution 5. Serologic immunoassay many infected individuals do not develop chronic lesions. The indicated method- ology for detection should be regarded as adequate for the protozoa and sufficient to support surveillance programs. Table 4 is a panel of classical bacterial pathogens of laboratory rats and mice. The indicated “standard” methods for detection should be regarded as adequate for surveillance programs in terms of accuracy of detection. Nonetheless, as time goes on, the polymerase chain reaction (PCR) is being more frequently applied for more faster and economical means of detection. The approach of using PCR for genome detection is particularly apt because the conventional methodology likewise requires detection of the microbes themselves, rather than immuno- serologic indicators of exposure (antibodies), which does form the primary detec- tion mode for the viruses. Note that serology is used, however, as the standard test mode of screening programs for rodent Mycoplasma infections and the cilia- associated respiratory bacillus for which serology forms a more useful screening device than cultural isolation. Bordetella was listed as a nod to tradition; cer- tainly it has not been a reported pathogen of laboratory rats and mice for many years, if ever. The most common and important viral pathogens of laboratory mice and rats are listed in Tables 5 and 6, respectively. It will be seen that with the exception 

STEVEN H. WEISBROTH 33 TABLE 4 Bacteria And Mycoplasmas—Diagnostic Alternatives for Rodent Comprehensive Health Surveillance Profiles Acceptable Alternate Bacteria and Mycoplasma Standard or Adjunctive Salmonella 1, 2 Streptobacillus moniliformis 1 Streptococcus pneumoniae 1 Streptococcus, B-hemolytic 1, 2 Cilia-associated resp. bacillus 4 3 Mycoplasma pulmonis 4 1, 3, 5, 6 Mycoplasma arthritidis 4 1, 3, 5, 6 Bordetella bronchiseptica 1, 2 Pasteurella pneumotropica 1, 2 5 Pseudomonas aeruginosa 1, 2 5 Citrobacter rodentium 1, 2 5 Klebsiella pneumoniae 1, 2 5 Klebsiella oxytoca 1, 2 5 Staphylococcus aureus 1, 2 5 Corynebacterium bovis (HAC) 1, 2 3, 5 1. Broth and agar media for primary isolation and semi-differentiation 2. Microtized media strips for biochemical profile and identification 3. Microscopic examination of histologic sections of relevant tissues 4. Serologic immunoassay 5. Polymerase chain reaction (PCR) 6. Corroborating gross pathology of Riley’s LDHV, testing strategy for the presence of these viruses is oriented to detection of antibodies engendered by viral infection. The assumption is (validly) made that viral infections reflect themselves by antibody production, and the presence or absence of the virus in the colony may be directly inferred by the presence or absence, respectively, of antibodies in the sample groups. With several exceptions, discussed below under Problematic Issues, the ELISA and immunofluorescent assay (IFA) are adequate methodology to support surveil- lance programs for all of the indicated agents (Lussier and others 1991). Indeed, the adequacy of these tests for surveillance programs has permitted essential eradication of all of these agents from quality breeding stocks in the United States. There are, however, diagnostic situations in which the actual presence or absence of the murine virus in test samples must be determined. These situations have classically been investigated by isolation in tissue culture or by the mouse (or rat) antibody production (MAP or RAP) test (Lussier and others 1991). Increasingly, because of the minimum 3 to 4 week time required to conduct a 

34 MICROBIAL AND PHENOTYPIC DEFINITION OF RATS AND MICE TABLE 5 Mouse Viruses—Diagnostic Alternatives for Rodent Comprehensive Health Surveillance Profiles Test Modesa Viruses Std Alt Other PVM Pneumonia Virus of Mice E I REO3 Respiratory Enteric Orphan III E I, H SEN Sendai Virus E I GD7 Theiler’s Murine Encephalomyelitis Virus (TMV) E I LCMV Lymphocytic Choriomeningitis E I HAN Hantaan Virus E I MVM Minute Virus of Mice I E H, PCR MPV Mouse Parvovirus I E H, PCR MHV Mouse Hepatitis Virus E I PCR KV Kilham’s Virus E H EDIM Epidemic Diarrhea of Infant Mice E I MAV Mouse Adenovirus E I ECTR Ectromelia Virus E I POLY Polyoma Virus E I MCMV Mouse Cytomegalovirus E I MTV Mouse Thymic Virus I LDHV Lactic Dehydrogenase Elevating Virus C aStd = Standard or Preferred, Alt = Alternate or Confirmatory, E = Enzyme Linked Immunosorbent Assay (ELISA), C = Biochemical Assay, I = Immunofluorescent Assay (IFA), H = Hemagglutination Inhibition (HAI) TABLE 6 Rat Viruses—Diagnostic Alternatives for Rodent Comprehensive Health Surveillance Profiles Test Modesa Viruses Std Alt Other PVM Pneumonia Virus of Mice E I REO3 Respiratory Enteric Orphan III E I SEN Sendai Virus E I GD7 Theiler’s Encephalomyelitis or TMI E I LCMV Lymphocytic Choriomeningitis E I HAN Hantaan Virus E I KRV Kilham’s Rat Virus E I H RPV Rat Parvovirus E I H TH1 Toolan’s H1 Virus E I SADV/RCV Sialodacryoadenitis Virus/Rat Corona Virus E I aStd = Standard or Preferred, Alt = Alternate or Confirmatory, E = Enzyme Linked Immunosorbent Assay (ELISA). I = Immunofluorescent Assay (IFA), H = Hemagglutination Inhibition (HAI) 

STEVEN H. WEISBROTH 35 MAP test, PCR applications are being employed for detection of viral contami- nants of biotechnical products and tissue culture cell lines, and for investigation of disease outbreaks. PCR applications presently have their greatest utility in testing for a single virus, and a growing body of literature attests to the use of PCR for murine virus detection. However, the versatility of the MAP test simul- taneously permits detection of all 16 to 18 agents of concern. Research is cur- rently under way to enable genomic detection of viral groups or multiplex tests to deal with this problem. Finally, it needs to be accepted that there is a residual group of agents not likely to be detected by traditional screening tests of nonlesioned rodent sample or sentinel groups. These agents, along with the Helicobacter species, share the feature of being difficult or impossible to isolate using routine microbiologic methods or to histopathologically demonstrate in the absence of lesions. They can be described as latent and clinically silent, and this term is used to list them in Table 7. Traditional means of detection for this important group have required first rendering the carrier hosts immunodeficient by chemical immunosuppres- sants (such as cortisone [and its synthetic derivatives]) or antimetabolites (such as cylosphosphamide) so as to encourage clinical recrudescence or florid expression of the pathogens, if present. This approach—the stress test—has been used particularly to reliably demonstrate closed rodent production units as free of the Tyzzer’s disease agent (Clostridium piliforme), Corynebacterium kutscheri, and Pneumocystis carinii (Weisbroth, 1995). It is with this group that PCR has had its greatest utility in rodent disease detection. Properly conducted, the stress test is arduous, takes 3 to 4 weeks to complete, and is expensive. By comparison, PCR can be done in 1 to 2 days, does not require ablation of the normative immune inhibition of pathogen populations, and is at least 1 to 2 logs more TABLE 7 Latent and Clinically Silent Agents Probably not Detected by Standard Methodology—Diagnostic Alternatives for Rodent Comprehensive Health Surveillance Profiles Acceptable Alternate Not Agents Preferred or Adjunctive Recommended Clostridium piliforme 1 2 3 Corynebacterium kutscheri 1 2, 3 Pneumocystis carinii 1 2 Helicobacter sp. 1 4, 3 1. Polymerase chain reaction (PCR) 2. Stress test with terminal samples processed by standard microbiologic and histopathologic methods 3. Serologic immunoassay 4. Specialized microbiologic isolation methods 

36 MICROBIAL AND PHENOTYPIC DEFINITION OF RATS AND MICE sensitive. PCR should be regarded as the preferred surveillance methodology for this group, with properly conducted stress tests also being considered as method- ologically adequate. PROBLEMATIC ISSUES I have indicated serology as “not recommended” for the diagnosis of Clostridium piliforme for several reasons. On the one hand, there is no basis in the literature for the use of serology as a screening device to detect nonlesioned carriers. More important, on the other hand, is the seeming predilection of rodents (and rabbits) to carry natural antibodies, ostensibly to commensal Clostridia, which induce low-level cross-reactive positivity to C. piliforme anti- gens in both ELISA and immunofluorescent assay test modes. Uncritical accep- tance of this type of positivity has led to interpretation of animal colony contami- nation with C. piliforme, when in fact this is not the case. Any provisional diagnosis of C. piliforme infection on the basis of serology alone needs support- ive corroboration by PCR, or histopathology, or both to confirm the diagnosis, or it is likely to be wrong. At present, a similar situation complicates serodiagnosis with a number of the murine viruses, in which the rodent host serum is tested with antigen to a murine virus, and antibodies to a different and presumptively human origin virus cause mainly low-titered cross-reactive positivity. Such is the situation with antibodies to REO III (REO3), Sendai (SEN), and SV5 in guinea pig serum from certain colonies, with reactivity to the GDVII (GD7) strain of Theiler’s encepha- lomyelitis virus (TMEV) in certain rat sera and, in the author’s opinion, antibod- ies to the agents now termed rat and mouse parvovirus, respectively, in rat and mouse serum. Thus, at present, the problematic issues with murine virus serol- ogy are not sensitivity, as they were with the earlier generation of complement fixation and hemagglutination inhibition methods, but rather, issues of specific- ity. Gradually it is becoming apparent that the problems of rodent virus infection are rather more complicated than simply being limited to indigenous rodent virus infection. These cross-reacting and confusing serologic reactions are nature’s way of telling us that these rodents are being exposed to other viruses, often under circumstances in which a contaminated host source other than the humans who come in contact with the animals is difficult to credibly posit. We can expect this issue (that is, the human-to-rodent interface) to remain a complicating issue in rodent diagnostics and rodent health until the animal care community comes to grips with regulating the interface better than we do at present. In concluding my remarks, I would like to address a number of points that relate to a global perspective in rodent health assessment. 

STEVEN H. WEISBROTH 37 1. Can we have a globally universal standard, in terms of panels of agents of which the rodents must be free to be accepted as the highest quality? In other words, could we truly have global harmonization of an infectious standard for each of the laboratory rodents? The answer is that between countries and regions, there would of course be substantial areas of overlap as there now is between the lists presented here as representing a US consensus and the published official FELASA lists. But equally true, there would probably always be discrepant agents of regional concern on the panels not regarded as significant elsewhere. As examples, the FELASA lists for rats and mice include Klossiella, Proteus, Leptospira, Escherichia coli, and Yersinia pseudotuberculosis—which you will note are not in the US panels presented here—whereas Pasteurella pneumotropica and Citrobacter rodentium are considered of importance in the United States but not by FELASA. Perhaps expert committees could be formed to critically exam- ine whether discrepant agents need to retained or could be safely deleted from the regional panels. 2. There is at present no means for ensuring a uniform standard for the potency, purity, or specificity of serologic test reagents, either of antigens or of positive control sera. Similarly, there is at present no means for ensuring avail- ability of testing reagents for diagnostic labs. There should be, perhaps at the ICLAS level, some objective means of comparing and evaluating adequacy and availability of testing reagents to remove this potential variable to comparability of surveillance programs from laboratory to laboratory, and from country to country. 3. The testing laboratories themselves, whether at the state, national, or private sector level, are not regulated and required to meet administered perfor- mance standards. Whether required for professional acceptability (dare I say accredited?) or by entirely voluntary participation, there should be some objec- tive ongoing assessment of laboratory performance. 4. Finally, I would like to give an opinion against patent protection for discoveries of new agents. An unfortunate trend has been the movement to cash in on diagnostic discoveries by patenting organisms as isolated from nature and subsequently characterized as pathogens. An example is Helicobacter hepaticus. We have seen that the patent on this agent has acted to restrict its availability for exploration and implementation of diagnostic tests to the detriment of improve- ments in rodent health surveillance. Presumably this was not the intention of the patent holders, but just as surely, that has been one of the net effects. The unintended effect of restricted availability should be noted and the impulse to patent such “discoveries” discouraged. 

38 MICROBIAL AND PHENOTYPIC DEFINITION OF RATS AND MICE REFERENCES Baker, D.G. 1998. Natural pathogens of laboratory mice, rats and rabbits and their effects on research. Clin. Microbiol. Rev. 11:231-266. ICLA [International Committee on Laboratory Animals]. 1971. Defining the Laboratory Animal. IV Symposium. National Academy of Sciences, Washington, DC. Kraft, V., A. A. Deeny, H. M. Blanchet, R. Boot, A. K. Hansen, G. Milite, J. R. Needham, W. Nicklas, A. Terrot, C. Rehbinder, Y. Richard, and G. De Vroey. 1994. Recommendations for the health monitoring of mouse, rat, hamster, guinea pig and rabbit breeding colonies. Lab. Anim. 28:1-12. Lussier, G. L., J. K. Davis, W. R. Shek, A. L. Smith, and G. Lussier. 1991. Detection methods for the identification of rodent viral and mycoplasmal infections. Lab. Anim. Sci. 41:199-225. Rehbinder, C., P. Baneux, D. Forbes, H. Van Herck, W. Nicklas, Z. Rugnya, and G. Winkler. 1996. FELASA recommendations for the health monitoring of mouse, rat, hamster, gerbil, guinea pig and rabbit experimental units. Lab. Anim. 30:193-208. Weisbroth, S. H. 1995. Pneumocystis carinii: Review of diagnostic issues in laboratory rodents. Lab. Anim. 24:36-40. Weisbroth, S. H. 1996. Post-indigenous disease: Changing concepts of disease in laboratory rodents. Lab. Anim. 25:25-33.