10 Central Nervous System
There are only two infectious agents that are considered primary pathogens of the central nervous system. Theiler's murine encephalomyelitis virus only rarely causes clinical disease in mice. Encephalitozoon cuniculi, a common pathogen in rabbits, has been reported in mice and rats only a few times and has not been observed to cause clinical disease in the latter species. Both agents have been encountered as contaminants of rodent cell cultures.
Low. Theiler's virus probably is rare as a natural infection of contemporary mice and rats.
Theiler's murine encephalomyelitis virus (TMEV) infection in mice resembles poliomyelitis virus infection in humans. However, there are few reported instances of natural disease caused by TMEV in mice (Theiler, 1934, 1937, 1941; Theiler and Gard, 1940a; Thompson et al., 1951) and only one in rats (McConnell et al., 1964). The major importance of TMEV is that experimental infections in mice provide valuable models of polio-
myelitis-like infection (Theiler, 1941; Gard, 1943) and virus-induced demyelinating disease (Dal Canto and Lipton, 1975; Lipton, 1975; Lipton and Dal Canto, 1979a,b).
TMEV is a RNA virus, family Picornaviridae, genus Enterovirus. Other members of this genus include polioviruses, coxsackieviruses, and enteroviruses. Strains of TMEV are divided into two groups based on pathogenicity for mice and in vitro growth characteristics. The small-plaque strains are Theiler's original (TO) and similar strains (DA, WW, TO4, Yale, and BeAn 8386). These produce small plaques in L929 cells, are relatively avirulent, cause chronic demyelinating disease in mice, do not hemagglutinate human erythrocytes, and are not readily cultivated in embryonating eggs. The large-plaque strains are FA, GDVII, and others that produce large plaques in L929 cells, are highly virulent for mice, may or may not hemagglutinate human erythrocytes, and grow readily in eggs (Lipton, 1980). The MHG strain was isolated from rats (Hemelt et al., 1974). All strains are antigenically related in the cross-neutralization, hemagglutination inhibition, and complement fixation tests (Hemelt et al., 1974).
The virus particles are naked icosahedral nucleocapsids, measuring 2030 nm in diameter, and have a single-stranded RNA genome. The virions contain four major structural polypeptides: VP1, VP2, and VP3, each with a molecular weight of 27,000-58,000, and VP4, with a molecular weight of 6,000. In the large-plaque strains, VP1 is slightly heavier than in the less virulent small-plaque strains (Stroop and Baringer, 1981; Downs, 1982, Matthews, 1982).
TMEV can be propagated in several continuous cell lines, but BHK-21 cells are most commonly used (Lipton, 1978a). TMEV is rapidly inactivated by heating at 50-55°C for 30 minutes and by treatment with 50% acetone or alcohol, but not by treatment with ether (Theiler and Gard, 1940a).
Laboratory mice (Downs, 1982) and rats (McConnell et al., 1964).
Natural TMEV infection has been reported most frequently in laboratory mice (Downs, 1982) and on one occasion in laboratory rats (McConnell et al., 1964), but not in wild Mus musculus (Downs, 1982). The prevalence of TMEV infection in mice has been reported to be 5% of commercial barrier breeding colonies (Lindsey et al., 1986a) and 62% of other mouse populations (Parker, 1980) in the United States. The prevalence of TMEV infec-
tion in laboratory rats is unknown; however, one survey found 44% of laboratory rats in the United States to be serologically positive (Parker, 1980). Unfortunately, these data probably overestimated the true prevalence of TMEV as they were obtained by use of the hemagglutination inhibition (HAI) test, known to give false positive results for TMEV (Kraft and Meyer, 1986; Van Der Logt, 1986).
In naturally infected mice the virus occurs at a low titer in the intestinal mucosa, intestinal contents, feces, and less frequently, in the mesenteric lymph nodes. Transmission is by the fecal-oral route. In infected colonies housed in ordinary shoe-box cages, the infection is usually acquired between 3 and 6 weeks of age and most animals eventually become infected. Virus has been demomstrated in the feces of an individual mouse for only 53 days and persistent infection of the intestinal tract is considered unlikely (Olitsky, 1940; Theiler and Gard, 1940b; Lipton and Rozhon, 1986).
Natural infections in mice are usually inapparent and, presumably, caused by less virulent wild-type strains of TMEV resembling TO. Clinical disease is a rare occurrence, appearing at a rate of only 1 in 4,000-10,000 infected animals. Affected mice have flaccid paralysis of one or both rear legs but otherwise appear normal. There is little or no mortality (Theiler, 1934, 1937; Olitsky, 1940; Theiler and Gard, 1940a).
One epizootic in mice has been reported to be due to a highly virulent strain of TMEV resembling GDVII (Thompson et al., 1951). Signs included circling, rolling, hyperexcitability, convulsions, tremors, weakness, or flaccid paralysis of hind legs in some mice, and high mortality (62% of 240 mice). The outbreak occurred in a mouse room where GDVII was being used experimentally and sanitation was admittedly poor.
The MHG strain of TMEV was originally isolated from three adult rats with natural disease consisting of circling, incoordination, tremors, and torticollis (McConnell et al., 1964).
Natural disease in mice results from the rare occurrence of viremia, i.e., dissemination of virus from the intestine to the spinal cord and brain. This occurs most frequently around 6-10 weeks of age. The predominant lesion is poliomyelitis, with necrosis and neuronophagia of ventral horn cells and nonsuppurative inflammation composed primarily of lymphocytes. Little if any secondary demyelination is seen in the natural disease. TMEV can be isolated from the lesions for at least 1 year (Theiler, 1934, 1937; Theiler and Gard, 1940a; Lipton and Dal Canto, 1979a).
A variety of neurologic disease models has been developed by using
different strains of TMEV, routes of virus inoculation, and strains of mice. These models resemble the natural disease very little because large doses of virus are usually inoculated intracerebrally. In general, the large-plaque strains (FA or GDVII) produce acute encephalitis and death in weanling mice only 4-5 days after inoculation (Theiler and Gard, 1940a). In contrast, the intracerebral inoculation of the small-plaque strains (TO. DA. WW, and others) causes an acute poliomyelitis after an incubation period of a few weeks, followed by persistent viral infection with varying degrees of chronic demyelination and remyelination after a few months, with the latter processes resembling multiple sclerosis in man (Dal Canto and Lipton, 1975, 1979, 1980, 1982; Lipton, 1975; Penney and Wolinsky, 1979; Brahic et al., 1981; Stroop et al., 1981; Rodriguez et al., 1983; Dal Canto and Barbano, 1984; Lipton et al., 1984). SJL/J, SWR, and DBA/2 mice are highly susceptible to chronic demyelination, whereas A, C57BL/6, C57BL/10 and DBA/1 mice are resistant (Lipton and Dal Canto, 1976b, 1979b; Lipton and Rozhon, 1986). Susceptibility to demyelination can be prevented by immunosuppressive regimens of cyclophosphamide or antilymphocyte serum (Lipton and Dal Canto, 1976a), is influenced by the H-2D region and is associated with TMEV-specific delayed-type hypersensitivity (Clatch et al., 1985), suggesting that it is immune mediated.
The HAI test with GDVII antigen and human type O erythrocytes has been the standard procedure for serologic screening of mouse stocks for many years (Lahelle and Horsfall, 1949; Fastier, 1950, 1951). It is essential that the HAI test be performed at 4°C to avoid false-positive results. Even so, the HAI probably still gives a significant proportion of false positives, and the enzyme-linked immunosorbent assay is now considered the test of choice (Kraft and Meyer, 1986; Van Der Loft, 1986). The complement fixation and serum neutralization tests may also be useful for some purposes, such as comparisons of the antigenic relatedness of TMEV strains (Downs, 1982). The mouse antibody production (MAP) test can be used for screening biologic materials for the presence of virus (Rowe et al., 1959, 1962).
Definitive diagnosis is usually made by isolation of virus from spinal cords or brains of mice with clinical disease, but it is also possible to isolate virus from the intestinal contents of mice with asymptomatic infection (Downs, 1982). The virus can be propagated in several continuous cell lines (Sturman and Tamm, 1966; Hemelt et al., 1974; Lipton, 1978a,b), but BHK-21 cells are most commonly used (Lipton and Dal Canto, 1979b).
The most practical method of control is to obtain mice from breeding populations that have been shown to be free of the infection by serologic
testing, followed by barrier maintenance and regular testing to reconfirm their TMEV-free status.
TMEV infection has been eliminated from valuable mouse stocks by foster nursing infant mice on TMEV-free mice or rats (von Magnus and von Magnus, 1948; Dean, 1951). Also, the isolation and quarantine of individual breeding pairs with subsequent selection of TMEV negative progeny have been successful eliminating the infection (Lipman et al., 1987). Cesarean derivation is effective but usually is not justified.
Interference with Research
There are a few examples in the literature in which indigenous TMEV infections in mice have interfered with studies of unrelated viruses in mice (Theiler and Gard, 1940a; Melnick and Riordan, 1947; Thompson et al., 1951).
Significance Encephalitozoon cuniculi has low significance for most studies. It is highly significant for studies involving passage of transplantable tumors and other materials in mice and rats.
1923: Levaditi et al. (1923a,b, 1924) discovered this organism in the brains of rabbits with encephalitis and named it Encephalitozoon cuniculi. In subsequent years it was found to be ubiquitous in laboratory rabbits and only occasionally present in laboratory rodents and other species (Shadduck and Pakes, 1971).
1953-1969: During this period E. cuniculi was incriminated as a cause of interference with research in which mice and rats were used. A syndrome characterized by hepatosplenomegaly and attributed to an ascites hepatitis agent was observed following the intraperitoneal passage of tissues in mice (Lackey et al., 1953; Morris et al., 1956; Jordan and Mirick, 1965a,b). It was later shown to be due to E. cuniculi (Nelson, 1962, 1967; Weiser, 1965; Arison et al., 1966). E. cuniculi was found to contaminate transplantable tumors resulting in altered experimental results in mice (Arison et al., 1966) and rats (Petri, 1965, 1966, 1967, 1968, 1969). Also, during this period the genus name was temporarily changed to Nosema (Shadduck and Pakes, 1971).
1969: Shadduck (1969) reported the first successful in vitro cultivation of E. cuniculi; he used rabbit choroid plexus cells.
1972: Hunt et al. (1972) demonstrated transplacental transmission of the organism in rabbits.
1973-1988: A large number of serologic tests were developed for diagnosis of E. cuniculi infection (see below).
Encephalitozoon cuniculi is a protozoan, order Microsporidia. Nosema cuniculi, and Nosema muris are synonyms for E. cuniculi. Mature spores are oval, 1.5 x 2.5 µm, occur intracellularly, and are not surrounded by a cyst wall. Spores are Gram positive, argyrophilic, variably acid fast, and have a periodic acid-Schiff-positive granule at one pole. Spores stain dark magenta or purple with Goodpasture's carbol fuchsin and stain poorly with hematoxylin and eosin (Perrin, 1943b; Weiser, 1965; Shadduck and Pakes, 1971; Wilson, 1979).
It has been grown successfully in tissue cultures of several cell lines (Shadduck, 1969; Bismanis, 1970; Cox and Pye, 1975; Waller, 1975; Shadduck and Geroulo, 1979).
Rabbits, mice, rats, hamsters, guinea pigs, dogs, nonhuman primates, humans, and many other mammals (Perrin, 1943a; Lainson et al., 1964; Petri, 1969; Shadduck and Pakes, 1971; Margileth et al., 1973; Shadduck et al., 1978; Zeman and Baskin, 1985).
E. cuniculi is considered ubiquitous in rabbits. Surveys of prevalence suggest that most rabbit colonies are infected (Koller, 1969; Flatt and Jackson, 1970; Cox and Pye; 1975; Waller, 1977; Lyngset, 1980) unless they have been specifically rederived to exclude this infection (Cox et al., 1977; Bywater and Kellett, 1978). Prevalence within infected colonies has ranged from 15 to 76% (Shadduck and Pakes, 1971). Rabbits undoubtedly provide the major source of infection for mice and rats in research facilities.
Prevalence of the organism in contemporary mouse and rat stocks is not known but is thought to be very low in comparison to that in rabbits. In a recent serologic survey in the United Kingdom, Gannon (1980a) found that 1 of 17 mouse colonies and 2 of 12 rat colonies were infected.
Transmission in most animals is primarily horizontal by the orofecal route (Wilson, 1979). Organisms are shed in the urine and ingested by
another host. Transmission in rabbits is both vertical (Hunt et al., 1972) and horizontal (Shadduck and Pakes, 1971; Cox et al., 1979). In mice, horizontal transmission is known to occur between cagemates, but transplacental transmission has not been demonstrated (Liu et al., 1988). Studies of transmission in rats have not been done.
Natural infections usually are inapparent.
In rats and rabbits the classic lesion of E. cuniculi infection in the brain is a meningoencephalitis with multifocal granulomatous inflammation. Activated macrophages form so-called glial nodules in response to the agent. These nodules can have necrotic centers or appear as solid sheets of cells. With special stains, the organisms can be seen in compact masses within cyst-like spaces in individual cells or scattered in the glial nodules. Varying numbers of lymphocytes and plasma cells are seen in the meninges and around vessels. The brain lesions in mice are similar, except for the lack of the granulomatous foci (Perrin, 1943b; Yost, 1958; Innes et al., 1962; Attwood and Sutton, 1965; Shadduck and Pakes, 1971; Shadduck et al., 1979; Majeed and Zubaidy, 1982).
The organism also occurs intracellularly in the renal tubular epithelium with or without the presence of an inflammatory response, and in the renal tubular lumens. In chronic infections focal destruction of tubules and replacement by fibrous connective tissue results in small pits on the cortical surface (Yost, 1958; Flatt and Jackson, 1970). Lesions in organs other than the kidney and brain are less consistent (Shadduck and Pakes, 1971).
Strain differences in susceptibility to experimental infection have been documented. Strains BALB/c, A/J, and SJL are resistant to experimental infection; strains C57BL/10, DBA/2, and AKR show intermediate susceptibility; and strains C57BL/6, DBA/1, and 129/J are very susceptible (Niederkorn et al., 1981). MRL/MpJ- lpr/lpr mice are no more susceptible than MRL/MpJ-+/+ mice (Liu and Shadduck, 1988).
Experimental infection of euthymic mice results in chronic nonlethal infection, whereas infection of athymic (nulnu) mice leads to death in a few weeks, suggesting that resistance is T-cell dependent (Gannon, 1980b; Niederkorn et al., 1981; Schmidt and Shadduck, 1983). Infection of immunocompetent mice results in transient augmentation of natural killer cell activity in spleen (Niederkorn et al., 1983) and lung (Niederkorn, 1985).
The intraperitoneal inoculation of E. cuniculi, as in the passage of contaminated transplantable tumors, results in ascites in mice (Nelson, 1962, 1967; Arison et al., 1966; Petri, 1969).
Disseminated disease caused by E. cuniculi has been reported in a severely immunocompromised human infant (Margileth et al., 1973).
Several serologic tests have been developed for diagnosis of the infection in rabbits, including an indirect immunofluorescent antibody (IFA) test (Chalupsky et al., 1973; Jackson et al., 1973; Cox and Pye, 1975; Cox and Gallichio, 1978), India ink immunoreaction (Waller, 1977; Kellett and Bywater, 1978), complement fixation (Wosu et al., 1977), immunoperoxidase (Gannon, 1978), indirect microagglutination (Shadduck and Geroulo, 1979), enzyme immunoassay (Cox et al., 1981), and enzyme-linked immunosorbent assay (Beckwith et al., 1988). However, only the IFA and immunoperoxidase tests have been used in surveying mouse and rat colonies (Gannon, 1980a). Serologic testing is superior to other methods for the screening of colonies.
Other methods used for diagnosis include detection of parasites in urine (Goodman and Garner, 1972; Pye and Cox, 1977), demonstration of typical lesions and organisms in tissue sections (Petri and Schiodt, 1966: Koller, 1969; Flatt and Jackson, 1970), and an intradermal skin test (Pakes et al., 1972). In rabbits, results of the intradermal test correlate well with the presence of brain lesions caused by the organism (Pakes et al., 1984b).
Although transplacental transmission occurs, serologic testing of adult animals with selection of E. cuniculi-free breeding stocks can be used successfully for eradicating the infection in rabbits (Cox et al., 1977; Bywater and Kellett, 1978), and possibly could be useful for doing so in mice and rats.
Cesarean derivation, barrier maintenance, and improved sanitation of mouse and rat stocks appear to have been effective in reducing the prevalence of E. cuniculi in these species in recent decades. Mice and rats should not be housed near rabbits unless the rabbits are known to be free of E. cuniculi.
Interference with Research
The histologic changes caused by E. cuniculi infection in the brain and kidneys can complicate the interpretation of lesions in studies requiring histopathology (Bender, 1925; Frenkel, 1956; Waksman and Adams, 1956; Innes et al., 1962; Attwood and Sutton, 1965; Howell and Edington, 1968; Shadduck and Pakes, 1971; Gannon, 1980a; Majeed and Zubaidy, 1982; Ansbacher et al., 1988).
E. cuniculi can contaminate transplantable tumors and alter host responses during their passage in mice (Arison et al., 1966), rats (Petri, 1965, 1966, 1967, 1968, 1969), and hamsters (Meiser et al., 1971).
Mice experimentally infected with E. cuniculi have reduced humoral antibody titers to sheep erythrocytes and reduced proliferative spleen cell responses to mitogens (Niederkorn et al., 1981; Didier and Shadduck, 1988), and altered natural killer cell activity (Niederkorn et al., 1983; Niederkorn, 1985). Altered humoral immune responses due to E. cuniculi infection also have been reported in rabbits (Cox, 1977; Cox and Gallichio, 1978; Waller et al., 1978).