Emerging (and Reemerging) Viruses of Laboratory Mice and Rats
Abigail L. Smith
Professor of Pathology, Loyola University Medical Center
Results of serologic tests performed to monitor laboratory rodents for infectious diseases are dramatically more accurate now than in the 1970s when serologic testing was largely limited to complement fixation and hemagglutination inhibition (HAI) tests. It was, in fact, the improvements in diagnostic testing that led to our appreciation of some “emerging” infections. The new generation of tests also revealed unexpectedly high prevalence of infection with some agents we thought were present at low levels in commercial and academic facilities. For example, many sera from mice that had sustained infection with mouse hepatitis virus yielded uninterpretable (anticomplementary) results in the complement fixation test. These were reported as “unsatisfactory” but were clearly positive when tested retrospectively by indirect fluorescent antibody (IFA) tests and enzyme immunoassays.
The agents discussed here were chosen for several reasons. They include (1) prevalence; (2) documented interference with biomedical research; (3) difficulty in eliminating and/or preventing infection, especially in populations of genetically altered rodents housed under crowded conditions; and (4) periodic reemergence from unexpected sources.
AGENTS OF CONCERN
The existence of at least one previously unrecognized parvovirus of mice was suspected in the early 1980s when the HAI assay was replaced in some laboratories by IFA tests that used minute virus of mice (MVM) as antigen. A
proportion of mouse sera that clearly reacted with MVM antigen in the IFA test yielded negative results in the HAI assay that was used for confirmation. Although some opined that the IFA test simply yielded false-positive reactions, others were suspicious that there was an agent (or agents) distinct from MVM circulating in laboratory mouse colonies. As we know, the latter was the case, and the discrepant results of the two tests were based on the fact that the IFA test permitted recognition of both structural and nonstructural proteins of parvoviruses. The nonstructural coding regions of MVM and the newly recognized mouse parvovirus (MPV) are essentially identical, whereas the structural regions (recognized by sera in the HAI assay) are quite divergent. It is likely that those differences in the structural region account for the fact that the humoral immune response protects only against homotypic parvovirus infections of mice (Hansen and others 1999).
The existence of a putative new parvovirus was supported by studies at Yale, where transmission within an enzootically infected breeding colony of mice was documented. Medium from cultured peripheral blood lymphocytes and explanted spleens of seropositive mice contained a substance that agglutinated mouse erythrocytes; however, the hemagglutination could not be inhibited by antibody to MVM, rat virus (RV), or H-1 virus (A.L. Smith, unpublished data). Several years later, a cellular immunology laboratory at the University of Chicago began to have difficulty maintaining cloned T cell lines. Some cell lines died suddenly, and others simply failed to thrive. The presence of aggregated mouse erythrocytes was perceived to be a reliable indicator of infection with a putative virus (McKisic and others 1993). Southern blot analysis revealed the presence of a parvovirus that was shown serologically to be distinct from both the prototype and immunosuppressive allotropic variants of MVM. Infected cultures responded poorly to specific antigen and to interleukin 2. The agent was presumably introduced into the laboratory by spleen cells used as feeders and/or substrates for producing growth factors in mixed lymphocyte cultures. That laboratory, as well as others concentrating on murine T cell immunology, had episodic difficulty maintaining T cell lines and clones. This may have been due to incomplete decontamination after infections were recognized—parvoviruses are notoriously stable in the environment. Retrospective serology confirmed that MPV has circulated in US mouse colonies at relatively high prevalence for more than 25 years (Jacoby and others 1996).
Mouse parvovirus has an ideal relationship with its natural host: Infected mice of all genotypes and ages so far tested remain clinically normal and manifest no pathologic changes. The virus does not follow all parvoviral dogma—for instance, adult mice are at least as susceptible as neonates to MPV infection (Smith and others 1993). This is in contrast to the situation with MVM and most other parvoviruses. The generally higher susceptibility of neonates is likely attributable to the requirement by parvoviruses of a cellular factor present during S phase for their own replication.
In situ hybridization studies with tissues from experimentally infected mice have revealed that MPV replicates preferentially in the small intestine and in lymphoid tissues. Viral DNA is apparently cleared from the intestine at some time after seroconversion but can be detected in lymphoid organs of experimentally infected mice for at least 9 weeks (Jacoby and others 1995). A single study has evaluated the serologic and virologic characteristics of mice from a colony enzootically infected with MPV (Shek and others 1998). Two-, 3- and 6-month-old BALB/c mice from that colony had MPV DNA in lymph nodes, spleen, and small intestine. Infectious virus was recovered from the same tissues of some of those mice. One-month-old mice from the colony were seropositive, and a homogenate of pooled small intestine contained infectious virus. Only the 1-month-old mice transmitted infection to cage contacts. The implication is that mice in an enzootically infected colony remain infected for at least 6 months and possibly for life; however, transmission studies suggested that older mice do not transmit infection to cage contacts, a small consolation for colony managers. We do not yet know whether intestinal infection can be reactivated by environmental factors or experimental manipulation, resulting in recurrent transmission.
The persistence of MPV in lymphoid tissue raised the possibility of aberrant or inappropriate immune responses against antigens to which the host might be exposed. Studies designed to address that possibility have revealed that MPV infection does modulate T cell effector function. Tumor allografts were rejected at an accelerated rate by MPV-infected mice, and T cells from infected mice that had rejected the tumors had diminished cytolytic capacity (McKisic and others 1995). In a separate series of studies, MPV potentiated the rejection of allogeneic skin grafts, but proliferation of alloantigen-reactive lymphocytes from graft-sensitized mice was reduced. Unexpectedly, MPV also induced rejection of syngeneic skin grafts, and T cells from infected, graft-sensitized mice lysed syngeneic target cells (McKisic and others 1998). Autoimmunity as a consequence of MPV infection is intriguing in view of recent reports suggesting that B19 virus may induce autoimmune disease in humans (Lunardi and others 1998; Vigeant and others 1994).
Ideally, biologic contaminant testing should be done on any cultured cells or tumors destined for use in laboratory animals. The policy is difficult to enforce, and some investigators must be convinced that compliance is in their best interest. Cells that will be injected into mice or rats must always be tested for rodent parvoviruses because of their affinity for rapidly dividing cells, such as tumor cells or lymphocytes. Additionally, there have been several instances in recent years of MVM infection of cells in large-scale production bioreactors (Chang and others 1997; Garnick 1996; A.L. Smith, unpublished data). Several laboratories have developed polymerase chain reaction methods for detection of rodent parvovirus contamination (Besselsen 1998; Besselsen and others 1995; Chang and others 1997; Garnick 1996; Riley and others 1999). This sensitive, rapid methodology should improve investigator compliance with institutional testing policies
because results can be available within 24 hours of sample submission. This contrasts with the 3- to 6-week turnaround interval for mouse (or rat) antibody production tests.
Rats are now known to harbor multiple parvoviruses. These agents seem to display more genetic heterogeneity than do parvoviruses of mice; however, this could simply be due to the fact that more rat parvoviruses have been isolated and characterized. The virus now called rat parvovirus (RPV)-1a shared only 82% amino acid identity in the NS coding region with the UMass strain of RV (Ball-Goodrich and others 1998). Four additional rat parvovirus isolates, two from wild rats, have been characterized at the molecular level. The two isolates from wild rats (RPV-2a) were identical, and there was 95% protein sequence similarity among those two viruses and two others (RPV-2b and RPV-2c) from geographically distinct colonies of laboratory rats (Wan and others 1999). However, nucleotide homology of the RPV-2a isolates, RPV-2b and RPV-2c, with the isolate characterized by Ball-Goodrich and others (1998) was only 73%.
A recent serologic study indicated that the prevalence of RPV in rats in Japan ranged from 13 to 22% (Ueno and others 1998). The same investigators reported that an uncharacterized isolate of RPV preferentially infected lymphoid tissue and was excreted in feces, urine and nasopharyngeal secretions (Ueno and others 1997). Viral DNA could be detected in lymphoid tissues for at least 24 weeks. RPV-1a preferentially infects lymphoid tissues and endothelium and, unlike isolates of RV, is enterotropic as well (Ball-Goodrich and others 1998).
Based on their affinity for lymphoid tissues, rat parvoviruses might be expected to modulate immune responses. The Kilham strain of RV induced T cell-dependent autoimmune diabetes in several strains of rats (Ellerman and others 1996). The UMass strain of RV infects both CD4+ and CD8+ T cells as well as B cells (McKisic and others 1995). T cells from infected rats proliferated poorly and had reduced cytolytic capacity. Nonlethal infection of a CD4+ T cell line resulted in reduced proliferation in response to antigen and interleukin 2 (McKisic and others 1995). The immunomodulatory properties of RPV isolates have not yet been explored.
Mouse Hepatitis Virus (MHV)
Mouse hepatitis virus is, in a sense, reemerging. The virus was successfully eliminated from many vivaria during the last 10 to 15 years. This was facilitated by the knowledge that infection is acute and self-limiting (Barthold and Smith 1987) and that cessation of breeding for a period of a few weeks would permit infection within a room to burn out (Weir and others 1987). Today we appear to be in the midst of a national epizootic of MHV, and recent finger pointing has
shed more heat than light on the problem. Why do we find ourselves in this frustrating situation? What follows is partly speculative and partly based on observations of vivaria at several large universities. First, it is almost impossible to eliminate MHV from every niche in an academic setting. There will always be one influential investigator who absolutely cannot, under any circumstances, stop his/her experiments long enough for the infection to run its course. If there is a single room, no matter how remote, housing MHV-infected mice, the risk of transmission exists. That risk can be traced to HVAC systems, fomites, and human tracking, among other possibilities. Yet influential investigators, the malfunctioning HVAC systems, fomites, and people who enter rooms out of order have always existed. What is so different now? The answer lies, I believe, in the nature of the contemporary murine host and the population density of that host. Genetically altered rodents are being developed and used in biomedical research at a staggering rate. Many of these animals sustain MHV infections that appear to be of longer duration than that seen in “normal” mice. In addition, it is not at all unusual to find mice housed at such high density that the filter tops, which formerly afforded protection against infections, are tipped and provide no physical barrier at all. Cage and rack manufacturers are responding to the problem with innovative designs for ventilation and maintenance of larger numbers of cages per rack. However, academic institutions are slow to respond with the resources needed to fix the problem on a more permanent basis—new facilities. It is probably not unreasonable to predict that any facility currently in the design phase will be filled to capacity on the opening day.
The risks to research done with MHV-infected mice have been well documented. MHV infection modulates both T cell and antigen-presenting cell function (de Souza and Smith 1991; de Souza and others 1991; Smith and others 1991). B cells of genetically susceptible strains of mice may become infected but remain viable and morphologically normal (de Souza and Smith 1991). More recent reports have shown that MHV can also noncytolytically infect embryonic stem cell lines derived from mice of many genetic backgrounds (Kyuwa 1997). This finding makes it imperative to use feeder cells from uninfected mice. Infected embryonic stem cells apparently do express normal levels of differentiation markers (Okumura and others 1996).
The causative agent of mousepox reemerges in the United States once or twice each decade, and it is included here because we need to be reminded periodically that this can happen. Earlier this year ectromelia virus was introduced into a single mouse room at Weill Medical College of Cornell University (Lipman and others 1999). The source of the infection was pooled mouse serum that had been prepared in China as a batch of at least 43 liters in early 1995. Given the volume, it is highly likely that there will be additional outbreaks
associated with this lot of serum. A similar outbreak occurred in 1995 at the Naval Medical Research Institute in Bethesda, Maryland (Dick and others 1996). Again, a single room was involved. Clinical signs were mild, the most common being conjunctivitis, and mortality was low. The source of infection was pooled mouse serum that had been prepared within the continental United States. The reagent was marketed for use in culture medium, not in vivo use. However, it is not unusual for cultured cells of several types (which may, themselves, be free of contaminating agents), grown in media supplemented with animal serum, to be injected into laboratory rodents. What is most surprising is that this was the sole outbreak to arise from use of that batch of pooled mouse serum. However, Jacoby and Lindsey (1998) reported that some respondents to a survey for infectious diseases in US rodent colonies indicated the presence of ectromelia virus infection in non-specific pathogen-free colonies of mice.
It is unlikely that any of the agents discussed here will be eradicated in the foreseeable future. In fact, several may increase in prevalence as more diverse transgenic and knock-out animals are developed. That some of these animals may have immune dysfunction is not always predictable based on the targeted genetic change. Compounding the problem is the issue of housing density, which contributes to transmission of infections among animals just as it does in human institutions such as day-care centers. Perhaps the best analogy in humans is adenovirus-associated acute respiratory disease in newly assembled military recruits housed under crowded conditions.
Will new or newly recognized infectious agents emerge among laboratory rodents? The answer is, emphatically, yes. This emergence is predicted by recent experience: Several rodent parvoviruses and Helicobacter species have been recognized in a relatively short span of time. Riley and colleagues have presented evidence of an infectious etiology for idiopathic lung lesions in rats (Riley and others 1999). Unquestionably, there will be more agents to keep veterinary microbiologists and virologists gainfully employed!
Ball-Goodrich, L.J., S.E. Leland, E.A. Johnson, F.X. Paturzo, and R.O. Jacoby. 1998. Rat parvovirus type 1: The prototype for a new rodent parvovirus serogroup. J. Virol. 72:3289-3299.
Barthold, S.W., and A.L. Smith. 1987. Responses of genetically susceptible and resistant mice to intranasal inoculation with mouse hepatitis virus JHM. Virus Res. 7:225-239.
Besselsen, D.G. 1998. Detection of rodent parvoviruses by PCR. Methods Mol. Biol. 92:31-37.
Besselsen, D.G., C.L. Besch-Williford, D.J. Pintel, C.L. Franklin, R.R. Hook, and L.K. Riley. 1995. Detection of H-1 parvovirus and Kilham rat virus by PCR. J. Clin. Microbiol. 33:1699-1703.
Chang, A., S. Havas, F. Borellini, J.M. Ostrove, and R.E. Bird. 1997. A rapid and simple procedure to detect the presence of MVM in conditioned cell fluids or culture media. Biologicals 25:415-419.
de Souza, M.S. and A.L. Smith. 1991. Characterization of accessory cell function during acute infection of BALB/cByJ mice with mouse hepatitis virus (MHV), strain JHM. Lab. Anim. Sci. 41:112-118.
de Souza, M.S., A.L. Smith, and K. Bottomly. 1991. Infection of BALB/cByJ mice with the JHM strain of mouse hepatitis virus alters in vitro splenic T cell proliferation and cytokine production . Lab. Anim. Sci. 41:99-105.
Dick, E.J., C.L. Kittell, H. Meyer, P.L. Farrar, S.L. Ropp, J.J. Esposito, R.M. Buller, H. Neubaner, Y.H. Kang, and A.E. McKee. 1996. Mousepox outbreak in a laboratory mouse colony. Lab. Anim. Sci. 46:602-611.
Ellerman, K.E., C.A. Richards, D.L. Guberski, W.R. Shek, and A.A. Like. 1996. Kilham rat triggers T-cell-dependent autoimmune diabetes in multiple strains of rat. Diabetes 45:557-562.
Garnick, R.L. 1996. Experience with viral contamination in cell culture. Dev. Biol. Stand. 88:49-56.
Hansen, G.M., F.X. Paturzo, and A.L. Smith. 1999. Humoral immunity and protection of mice challenged with homotypic or heterotypic parvovirus. Lab. Anim. Sci. 49:380-384.
Jacoby, R.O., L.J. Ball-Goodrich, D.G. Besselsen, M.D. McKisic, L.K. Riley, and A.L. Smith. 1996. Rodent parvovirus infections. Lab. Anim. Sci. 46:370-380.
Jacoby, R.O., E.A. Johnson, L.J. Ball-Goodrich, A.L. Smith, and M.D. McKisic. 1995. Characterization of mouse parvovirus infection by in situ hybridization . J. Virol. 69:3915-3919.
Jacoby, R.O. and J.R. Lindsey. 1998. Risks of infection among laboratory rats and mice at major biomedical research institutions. ILAR J. 39:266-271.
Kyuwa, S. 1997. Replication of murine coronaviruses in mouse embryonic stem cell lines in vitro. Exp. Anim. 46:311-313.
Lipman, N.S., H. Nguyen, and S. Perkins. 1999. Threat to U.S. colonies. Science 284:1123.
Lunardi, C., M. Tiso, L. Borgato, L. Nanni, R. Millo, G. De Sandre, A.B. Severi, and A. Puccetti. 1998. Chronic parvovirus B19 infection induces the production of anti-virus antibodies with autoantigen binding properties. Eur. J. Immunol. 28:936-948.
McKisic, M.D., D.W. Lancki, G. Otto, P. Padrid, S. Snook, D.C. Cronin, P.D. Lohmar, T. Wong, and F.W. Fitch. 1993. Identification and propagation of a putative immunosuppressive orphan parvovirus in cloned T cells. J. Immunol. 150:419-428.
McKisic, M.D., J.D. Macy, M.L. Delano, R.O. Jacoby, F.X. Paturzo, and A.L. Smith. 1998. Mouse parvovirus potentiates allogeneic skin graft rejection and induces syngeneic graft rejection. Transplantation 65:1436-1446.
McKisic, M.D., F.X. Paturzo, D.J. Gaertner, R.O. Jacoby, and A.L. Smith. 1995. A nonlethal rat parvovirus infection suppresses rat T lymphocyte effector functions. J. Immunol. 155:3979-3986.
Okumura, A., K. Machii, S. Azuma, Y. Toyoda, and S. Kyuwa. 1996. Maintenance of pluripotency in mouse embryonic stem cells persistently infected with murine coronavirus. J. Virol. 70:4146-4149.
Riley, L.K., A.J. Carty, and C.L. Besch-Williford. 1999. PCR-based testing as an alternative to MAP testing. Lab. Aaim. Sci. 49:443.
Riley, L.K., J.H. Simmons, G. Purdy, R.S. Livingston, C.L. Franklin, C.L. Besch-Williford, and R.J. Russell. 1999. Research update: Idiopathic lung lesions in rats. ACLAD Newslet. 20:9-11.
Shek, W.R., F.X. Paturzo, E.A. Johnson, G.M. Hansen, and A.L. Smith. 1998. Characterization of mouse parvovirus infection among BALB/c mice from an enzootically infected colony. Lab. Anim. Sci. 48:294-297.
Smith, A.L., R.O. Jacoby, E.A. Johnson, F. Paturzo, and P.N. Bhatt. 1993. In vivo studies with an “orphan” parvovirus of mice. Lab. Anim. Sci. 43:175-182.
Smith, A.L., D.F. Winograd, and M.S. de Souza. 1991. In vitro splenic T cell responses of diverse mouse genotypes after oronasal exposure to mouse hepatitis virus, strain JHM. Lab. Anim. Sci. 41:106-111.
Ueno, Y., M. Iwama, T. Oshima, F. Sugiyama, A. Takakura, T. Itoh, and K. Yagami. 1998. Prevalence of “orphan” parvovirus infections of mice and rats. Exp. Anim. 47:207-210.
Ueno, Y., F. Sugiyama, Y. Sugiyama, K. Ohsawa, H. Sato, and K. Yagami. 1997. Epidemiologic characterization of newly recognized rat parvovirus, “rat orphan parvovirus.” J. Vet. Med. Sci. 59:265-269.
Vigeant, P., H.A. Menard, and G. Boire. 1994. Chronic modulation of the autoimmune response following parvovirus B19 infection. J. Rheumatol. 21:1165-1167.
Wan, C.-H., D.J. Pintel, and L.K. Riley. 1999. Molecular characterization of four newly identified rat parvoviruses (RPV). Lab. Anim. Sci. 49:444.
Weir, E.C., P.N. Bhatt, S.W. Barthold, G.A. Cameron, and P.A. Simack. 1987. Elimination of mouse hepatitis virus from a breeding colony by temporary cessation of breeding. Lab. Anim. Sci. 37:455-458.