9 Hemopoietic System
The agents included in this section constitute a diverse assortment with very different host-parasite relationships. However, they do share some degree of commonality in pathogenetic mechanisms in that they all appear to have cell populations in the hemopoietic system as their chief target tissues.
Although lymphocytic choriomeningitis virus (LCMV) has traditionally been considered mainly a pathogen of the central nervous system, viremia and the immunologic responses of the host have central roles in pathogenesis, thus justifying its recognition as a pathogen of the hemopoietic system. The lactic dehydrogenase-elevating virus (LDV) causes persistent viremia, apparently replicates only in a small subset of macrophages, and through unknown mechanisms alters numerous immunologic functions and decreases the plasma clearance of lactic dehydrogenase isozyme V. Both Haemobartonella muris and Eperythrozoon coccoides are blood parasites. The murine leukemia viruses (MuLVs) are endogenous viruses found in all host cells of mice but cause hemopoietic neoplasms under appropriate conditions.
All of these agents have special importance in experimental oncology. MuLV-infected animals are frequently used as animal models for the study of viral carcinogenesis. The others, LCMV, LDV, H. muris, and E. coccoides, are notorious as inadvertent contaminants of transplantable tumors and other test systems, in which they frequently alter research results. In addition, LCMV infection has importance as a zoonosis.
Lymphocytic Choriomeningitis Virus
Lymphocytic choriomeningitis virus (LCMV) is a highly significant zoonotic infection of laboratory personnel who work with transplantable rodent tumors (particularly in hamsters) and rodent cell lines (Biggar et al., 1976; Pike, 1979), and of owners of pet hamsters (Parker et al., 1976). The virus also has importance as an unwanted contaminant that can alter research results of in vitro and in vivo rodent test systems (see below) and for the experimental study of virus-host interactions (Lehmann-Grube et al., 1983).
Lymphocytic choriomeningitis virus (LCMV) has been recognized as an important zoonotic agent and indigenous rodent pathogen for about five decades (Hotchin, 1971b). Traditionally, mice have been considered the primary source of LCMV infection. However, since 1960 three epidemics of LCMV infection involving at least 236 human cases have occurred in the United States, and all have been associated with Syrian hamsters, either as laboratory animals bearing transplantable tumors or as pets (Gregg, 1975). Because of the unique host-parasite relationship of LCMV infection in mice, experimental infections in mice have been used extensively as models for the investigation of such diverse phenomena as virus-specific immunological tolerance, virus-induced T-cell-mediated immunopathology, virus-induced immune complex disease, in vivo viral interference, and activation of natural killer cells (Lehmann-Grube et al., 1983).
LCMV is an RNA virus, family Arenaviridae, genus Arenavirus. LCMV is the type species of the genus. Other members of this genus include Lassa virus and the Tacaribe Complex (Amapari, Junin, Latino, Machupo, Parana, Pichinde, Tacaribe, and Tamiami Viruses). The LCMV strains used most widely for experimental purposes are WE, E-350, and CA 1371 (LehmannGrube, 1982; Matthews, 1982).
LCMV and other members of the genus Arenavirus have unique ultrastructural characteristics. The virions are pleomorphic, have a diameter of 50-300 nm (mean, 110-130 nm), and consist of a membranous envelope with surface projections surrounding an interior containing granules (host ribosomes that measure 20-25 nm in diameter) instead of a defined core. Virus particles bud from plasma membranes of infected cells and can form large intracytoplasmic inclusions (Murphy and Whitfield, 1975; Rawls and Buchmeier, 1975; Rawls and Leung, 1979).
LCMV and other members of the genus Arenavirus are highly sensitive to lipid solvents, detergents, and disinfectants such as formaldehyde. Infectivity is rapidly lost at pH values below 5.5 and above 8.5 (Biggar et al., 1976; Rawls and Leung, 1979).
LCMV can be propagated in a wide range of mammalian cells, but BHK21, L, and Vero cells are most commonly used (Rawls and Leung, 1979; Matthews, 1982).
Wild mice (Mus musculus) are the principal reservoir hosts, but laboratory mice and Syrian hamsters also serve as important natural hosts. Humans, monkeys, dogs, rabbits, guinea pigs, rats, and chickens are also susceptible (Wenner, 1948; Hotchin, 1971b; Ackerman, 1973; Forster and Wachendorfer, 1973; Gregg, 1975; Skinner et al., 1977; Smith et al., 1984b). LCMV also utilizes numerous cell lines as laboratory hosts, including transplantable tumor lines (Biggar et al., 1976) and tissue culture cell lines (van der Zeijst et al., 1983).
Infections of LCMV are generally thought to be very common in wild M. musculus (Ackerman, 1973). Of 1,795 house mice trapped in West Germany, 65 (3.6%) were found to be infected (Ackerman, 1973). Following the discovery of LCMV in laboratory mice at a research institute in England, the infection was diagnosed in 51 (67%) of 76 wild mice trapped in the vicinity (Skinner et al., 1977).
Natural LCMV infections have been reported in only about five colonies of laboratory mice (Traub, 1935; Findlay et al., 1936; Lepine and Sautter, 1936; Skinner and Knight, 1969; Smith et al., 1984b). Sato and Miyata (1986) surveyed laboratory mice in Japan for anti-LCMV antibodies using an indirect fluorescent antibody (IFA) test, and found seropositives in 3 (2.2%) of 152 specific-pathogen-free mice and 30 (5.6%) of 539 conventional mice.
Although hamsters infected with LCMV were found to be widely dispersed in the United States in recent years, those infections were traced to a single hamster breeder and a nearby laboratory supplying investigators with transplantable hamster tumors (Gregg, 1975). One can only conclude that past health monitoring procedures for hamsters and transplantable hamster tumors have been inadequate and that the spread of LCMV infection can be extremely insidious.
Only infected mice and hamsters are known to transmit the virus. Both
species can have chronic infections, with high concentrations of virus shed in urine, saliva, and milk. The portals of entry are probably mucus membranes and broken skin. Vertical (transovarian and/or transuterine) transmission is known to occur in mice and is considered 100% efficient for mice born to infected dams. Vertical transmission is also thought to occur in hamsters but has not been definitively proved. Once introduced into a population of mice, the infection can spread to all members of that population (Traub, 1939; Skinner and Knight, 1971; Parker et al., 1976; Lehmann-Grube. 1982).
Clinical signs in mice vary greatly, depending on the strain of virus, strain of mouse, and age of mice at the time of infection, but the following two forms of natural LCMV infection are generally recognized:
- Persistent tolerant infection. This form results when infection is acquired in utero or within a few days after birth. There is lifelong viremia and shedding of virus. Transient runting can occur during the first 3 weeks of life. Thereafter, the mice appear normal but can have modest growth retardation. At 7-10 months of age, immune complex glomerulonephritis occurs and is associated with emaciation, ruffled fur, hunched posture, ascites, and some deaths (Hotchin, 1962; Pollard and Sharon, 1973; Traub, 1973; Lehmann-Grube, 1982).
- Nontolerant (acute) infection. This form occurs when infection is acquired after the first week of life (after the development of immunocompetence). Viremia occurs, but there is no shedding of virus. The outcome is either death within a few days or weeks, or recovery with elimination of the virus. Most of the available data on this form of infection have been obtained from experimental inoculation of the virus intracerebrally, subcutaneously, intraperitoneally, or intravenously (not the normal routes). However, depending on the mouse strain, natural infections of LCMV in adult mice can range from inapparent to severe disease with high mortality. Considering that few natural outbreaks have been reported in mice, asymptomatic infections in adults may be the most common (Hotchin, 1962; Hotchin and Benson, 1963; Skinner et al., 1977).
Natural infections in hamsters are generally considered subclinical (Forster and Wachendorfer, 1973; Bowen et al., 1975). However, Parker et al. (1976) studied experimental LCMV infections in hamsters and reported that hamsters with persistent viremia developed progressive glomerulonephritis and had reduced litter sizes. They also observed runting in some congenitally infected animals.
In humans, the usual clinical manifestations are those of flu-like disease, with fever, headache, myalgia, nausea, vomiting, sore throat, and photophobia
being the major symptoms. Occasionally, rash, diarrhea, cough, lymphadenopathy, orchitis, delirium, or amnesia occurs. Unless physicians are alerted to the possibility of LCMV infection, the disease is frequently diagnosed as influenza, mononucleosis, herpes encephalitis, or tuberculous meningitis (Biggar et al., 1976).
Experimentally, the type and degree of LCMV-induced disease in mice have been shown to be greatly influenced by the virus strain, dose, and route of inoculation; mouse strain and age; and other factors (Parker et al., 1976; Lehmann-Grube, 1982). However, there are two types of natural disease that are dependent on the age of the mouse at the time of infection:
- Persistent tolerant infection. Infection occurs in utero or shortly after birth; lifelong viremia and virus shedding occur. These mice are not tolerant in the classical sense but have a kind of split tolerance in that T-cell activity is suppressed while B-cell activity is not. Specifically, there is diminution of LCMV-specific, major histocompatibility complex class I restricted cytotoxic T-lymphocyte activity, but unabated generation of LCMV-specific B-cell responses with production of high titers of antibodies to all LCMV polypeptides (Francis et al., 1987). Infectious virus circulates bound to LCMV-specific IgG and complement. These complexes accumulate in the renal glomeruli, choroid plexus, and, to a lesser degree, in synovial membranes, in blood vessel walls, and beneath the epidermis of the skin to cause late-onset disease that becomes apparent clinically around 7-10 months of age (Hotchin and Collins, 1964; Oldstone and Dixon, 1967, 1969, 1970; Kajima and Pollard, 1970; Buchmeier and Oldstone, 1978; Oldstone et al., 1980, 1983; Lehmann-Grube et al., 1983; Moskophidis and Lehmann-Grube, 1984). T-lymphocytes of the helper subset (Thy-1.2+, Lyt-2-, L3T4+) are persistently infected with LCMV (Tishon et al., 1988). There is generalized lymphoid hyperplasia and perivascular accumulation of lymphocytes and plasma cells in all visceral organs (Pollard and Sharon, 1969, 1973). Interferon is thought to have a central role in the development of immune complex disease (Riviere et al., 1980; Ronco et al., 1980; Jacobson et al., 1981; Saron et al., 1982; Woodrow et al., 1982).
- Nontolerant (acute) infection. Infection is acquired after the development of immunocompetence. Viremia occurs but there is no shedding of virus. If the mouse recovers from the acute disease, virus is eliminated from the blood and tissues in weeks or months (Hotchin, 1971a). Almost all of the data on this type of infection have been obtained from mice infected experimentally by the intracerebral, intraperitoneal, intravenous, or subcutaneous route. After intracerebral inoculation of LCMV, the predominant lesions are meningoencephalitis and hepatitis with large infiltrates of
- lymphocytes (Lillie and Armstrong, 1945; Collins et al., 1961; Rodriguez et al., 1983). When the virus is given by the other routes, there is multifocal hepatic necrosis and generalized lymphocytolysis of T and B cells with fibrinoid necrosis by four to five days. In survivors, regeneration of lymphoid organs commences on about day 9 (Mims and Tosolini, 1969; Lehmann-Grube and Lohler, 1981; Lohler and Lehmann-Grube, 1981). Both the morphologic lesions and elimination of the virus in the nontolerant infection are due to cell-mediated immune responses involving H-2-restricted cytotoxic T lymphocytes (Zinkernagel and Doherty, 1973, 1975, 1979; Doherty et al., 1974) and, possibly, natural killer cells (Welsh, 1978; Welsh and Doe, 1980). T-lymphocytes of the cytotoxic-suppressive subset (Lyt-2+) are thought to be particularly important in elimination of the virus (Moskophidis et al., 1987). LCMV-infected mice can be protected from disease by numerous immunosuppressive regimens (Lehmann-Grube, 1982). Athymic (nu/nu) mice inoculated with the virus at 3 to 6 weeks of age do not develop disease but have persistent viremia (Christoffersen et al., 1976; Christoffersen and Bro-Jorgensen, 1977; Ronco et al., 1981).
LCMV infection in hamsters appears to have a pathogenesis similar to that in mice, but far less is known about the infection in hamsters. Hamsters with congenital LCMV infection and hamsters inoculated with the virus as newborns develop persistent viremia and viruria through three months of age. After that time half of them clear the virus and remain healthy; the others maintain the viremia and viruria and develop chronic glomerulonephritis and generalized chronic vasculitis presumably due to immune complex disease. Adult hamsters inoculated with LCMV develop viremia and viruria but clear the virus in three to six months without becoming diseased (Parker et al., 1976).
Serologic methods are the most practical procedures for routine diagnosis, but bleeding and processing of blood from an animal suspected of LCMV infection should be done with care because of the likelihood of viremia. The methods of choice are the IFA, micro plaque-reduction test for neutralizing antibody, and enzyme-linked immunosorbent assay. The IFA is particularly useful for rapid diagnosis early in the course of infection, while the micro plaque-reduction test is considered best for chronic infection. The complement fixation test is considered relatively insensitive and is not recommended (Hotchin and Sikora, 1975; Kimmig and Lehmann-Grube, 1979; Lehmann-Grube et al., 1979; Ivanov et al., 1981; Smith et al., 1984b).
The mouse antibody production (MAP) test can be used in testing transplantable tumors and other biologic materials for contamination with LCMV (Rowe et al., 1959a, 1962). Alternatively, virus isolations in tissue
cultures followed by detection of LCMV antigen by using immunofluorescence (Hotchin and Sikora, 1975), or direct examination by immunofluorescence methods of tissues suspected of having LCMV infection (Wilsnack and Rowe, 1964) can be useful in selected instances.
The most practical method of control is to obtain mice and hamsters only from populations shown by regular health surveillance testing to be free of LCMV and to maintain them in a barrier facility that excludes wild rodents. Such stocks should also be retested at regular intervals to reconfirm their LCMV-free status. In addition, all biologic materials such as transplantable tumors coming into the facility must be pretested and shown to be free of the virus before experimental use. Periodic testing of animals receiving such materials is also desirable (Rowe et al., 1962; Collins and Parker, 1972; Bowen et al., 1975).
Once LCMV infection is diagnosed in mice, hamsters, a transplantable tumor, or other biologic materials, the entire stock should be destroyed and incinerated. Animal cages and other equipment should be removed and autoclaved. Animal rooms should be fumigated with either formalin [40% formaldehyde in water, sprayed on all room surfaces at 35.7 ml/m3 (1 ml/ ft3)] or paraformaldehyde [10.7 g/m3 (0.3 g/ft3) vaporized in a high-temperature silicone fluid at 96°C (205°F)] and allowed to remain vacant for 7-10 days (Biggar et al., 1976).
Cesarean derivation of animal stocks is of no value because of transovarian or transuterine infection (Hotchin, 1962; Skinner and Knight, 1971; Parker et al., 1976; Lehmann-Grube, 1982).
Interference with Research
LCMV infection is an important zoonotic infection that can result in fatality or have serious complications such as meningitis, orchitis, arthritis, and alopecia (Lewis et al., 1965; Baum et al., 1966; Hirsch et al., 1974; Bowen et al., 1975; Biggar et al., 1976; Pike, 1979).
LCMV has been a frequent contaminant of biologic materials used in research, including the following:
- Transplantable tumors of mice (DeBuryn, 1949; Stewart and Haas, 1956; Haas, 1960; Molomut and Padnos, 1965; Molomut et al., 1965; Collins and Parker, 1972; Bhatt et al., 1986), hamsters (Lewis et al., 1965; Baum et al., 1966; Bowen et al., 1975; Gregg, 1975; Biggar et al., 1976), and guinea pigs (Nadel and Haas, 1955; Jungeblut and Kodza, 1963).
- Tissue culture cell lines (Lehmann-Grube et al., 1969; van der Zeijst et al., 1983).
- Virus stocks, including leukemia viruses (Collins and Parker, 1972), distemper virus (Dalldorf, 1939), rabies virus (Wiktor et al., 1965), and mouse poliomyelitis virus (Wenner, 1948).
- Toxoplasma gondii sublines (Grimwood, 1985).
LCMV infection has an inhibitory effect on tumor induction due to polyoma virus (Hotchin, 1962), Rauscher virus (Youn and Barski, 1966), and mammary tumor virus (Padnos and Molomut, 1973) in mice, and transplantable leukemias in guinea pigs (Nadel and Haas, 1955; Jungeblut and Kodza, 1963) and mice (Padnos and Molomut, 1973).
LCMV infection in mice causes induction of natural killer cell activity early in the infection and proliferation of virus-specific cytotoxic T lymphocytes in chronic infection (Zinkernagel and Doherty, 1975, 1979: Welsh, 1978; Welsh and Doe, 1980; Pfau et al., 1982).
LCMV infection causes severe depression of humoral and/or cellular immunity in mice (Mims and Wainwright, 1968; Lehmann-Grube et al., 1972; Bro-Jorgensen and Volkert, 1974; Bro-Jorgensen et al., 1975; Guttler et al., 1975; Thomsen et al., 1982; Wu-Hsieh et al., 1988).
LCMV infection delays rejection of skin (Lehmann-Grube et al., 1972) and tumor (Guttler et al., 1975) allografts.
LCMV infection increases susceptibility of mice to ectromelia virus (Mims and Wainwright, 1968) or Eperythrozoon coccoides (Seamer et al., 1961) infection.
LCMV infection increases susceptibility of mice to bacterial endotoxin (Hotchin. 1962; Barlow, 1964).
LCMV infection increases susceptibility of mice to x-irradiation (Bro-Jorgensen and Volkert, 1972).
LCMV infection abrogates the naturally occurring insulin-dependent diabetes mellitus of BB rats, presumably by suppressing autoimmune mechanisms causing the disease (Dyrberg et al., 1988; Schwimmbeck et al., 1988).
Lactic Dehydrogenase-Elevating Virus
Lactic dehydrogenase-elevating virus is highly significant for research involving transplantable tumors, viral oncology, immunology, and serial passage of infectious agents in mice. It has low significance for many studies using mice.
In studies to develop a method for early detection of cancer, Riley and Wroblewski (1960) found that a 5- to 10-fold increase in serum lactate
dehydrogenase (LDH) occurred in mice following inoculation with Ehrlich carcinoma cells. Riley et al. (1960) demonstrated that this effect was due to a transmissible agent, lactic dehydrogenase-elevating virus (LDV), which was associated with his transplantable mouse tumors. Subsequently, Riley (1968) showed that more than 50 transplantable mouse tumors were contaminated with LDV, and evidence from many other laboratories (Notkins, 1965; Riley, 1974; Rowson and Mahy, 1975) further incriminated this virus as a major variable in tumor immunobiology.
LDV is an RNA virus, family Togaviridae, presently assigned to an unnamed genus that includes hog cholera, bovine diarrhea, equine arteritis, and simian hemorrhagic fever viruses (Brinton, 1982; Matthews, 1982; Rowson and Mahy, 1985). Virions are enveloped and average 50-55 nm in diameter. The nucleocapsid is approximately 30-35 nm in diameter. Maturation of virions occurs by budding from cytoplasm into intracytoplasmic vesicles (Brinton, 1982; Rowson and Mahy, 1985).
Antigenic diversity between the various isolates of LDV is poorly understood because the virus-antibody complexes that are present in infected mice interfere with conventional serum neutralization tests. Limited studies with heterologous antibody produced in rats or rabbits suggest that there are at least two serologically distinct strains (Bailey et al., 1965b; Cafruny and Plagemann, 1982; Rowson and Mahy, 1985).
LDV can be propagated in primary tissue cultures of mouse origin, including spleen, bone marrow, embryo fibroblast, and peritoneal exudate cells. However, virus production in such cultures generally declines after the first week, an effect thought to be due to the depletion of permissive macrophages. This effect can be largely eliminated by adding 10% L-cell-conditioned medium which provides a macrophage growth factor (Brinton, 1982; Rowson and Mahy, 1985).
LDV is inactivated by lipid solvents, detergents, and acid pH. The virus can be stored indefinitely in plasma at -70°C, but not at 4°C. Virus-infected plasma or feces retains infectivity for only about 24 hours at room temperature (Brinton, 1982).
Mice (Mus musculus). The virus does not replicate in rats, hamsters, guinea pigs, rabbits, or in cell cultures from them (Notkins, 1965; Brinton, 1982; Rowson and Mahy, 1985).
LDV has been isolated from wild mice in Australia (Pope, 1961), Europe (Rowson, 1963; Georgii and Kirschenhofer, 1965; Field and Adams, 1968), and the United States (Pope and Rowe, 1964). Wild mice presumably serve as reservoir hosts.
LDV infection is not likely to be seen in breeding colonies of laboratory mice, but is very likely to occur in mice used in certain types of experiments if appropriate preventive measures are not followed. Transmission occurs most readily during experimental procedures such as mouse-to-mouse passage of contaminated tumors, cells, or serum, or use of the same needle to inoculate multiple mice. Experimentally, any parenteral route is effective (Notkins, 1965).
Although mice infected with LDV shed the virus in feces, urine, saliva, and milk, the virus titer in these excretions declines sufficiently after the first week of infection that the risk of transmission to other mice is subsequently relatively low. A similar pattern also holds for transplacental transmission. Dams infected during gestation can have a high percentage of infected progeny, but dams infected one week before mating or a few days after parturition have relatively few infected progeny. Fighting (bite wounds) increases transmission between cage mates (Notkins and Scheele, 1963; Plagemann et al., 1963; Crispens, 1964; Notkins, 1965).
LDV infection results in lifelong viremia in which the virus is complexed to antiviral antibody, but clinical signs do not occur (Rowson and Mahy, 1985). The exception to this general rule is the flaccid paralysis seen in C58 (strain designation incompletely given) and AKR strain mice with age-dependent polioencephalomyelitis caused by LDV at 5 months of age or older (see below).
After infection the LDV titer in the mouse's serum reaches 1010-1011 median infectious doses (ID50) 12-14 hours after infection, drops to 107 ID50/ml by 72-96 hours, and drops further to 105 ID50/ml by about 2 weeks, when the titer stabilizes for life. The virus replicates in macrophages so that virus titers in spleen, lymph nodes, liver, and thymus are similar to those in serum (Notkins, 1965).
The activity of plasma LDH begins to rise about 24 hours after infection, peaks at an 8- to 11-fold increase above normal at 72-96 hours after infection,
and then gradually declines over the next three months but remains significantly elevated for life. SJL/J mice are unique in that they experience a 15- to 20-fold increase in serum LDH, a recessive trait of this strain (Crispens, 1971, 1972; Inada and Mims, 1987). The increase in LDH is due to decreased clearance of only one of five isozymes of LDH, LDH V (Plagemann et al., 1963; Warnock, 1964). Several other plasma enzymes are increased in activity but to a lesser degree than LDH (Notkins, 1965; Brinton, 1982).
The virus replicates only in a small subpopulation of macrophages, the specific identity of which remains in question. Initially, considerable evidence seemed to implicate the Ia positive macrophage as the susceptible cell (Inada and Mims, 1984, 1985a,b, 1987). More recently, it has been shown that macrophages with a trypsin-sensitive receptor (Ia antigen is not trypsin-sensitive) are the permissive subpopulation (Kowalchyk and Plagemann, 1985; Buxton et al., 1988). SJL/J mice apparently have more of these cells than other strains of mice (Inada and Mims, 1987). Cytopathic effects have been seen in vitro in these cells. The virus replicates for one cell cycle only, unless the specific subpopulation of macrophages is replenished. After infection these cells have impaired antigen-presenting capacity and do not trigger memory T cells (Isakov et al., 1982a,b; Stueckemann et al., 1982a,b). Acute infection also causes rapid, transient interferon production (Evans and Riley, 1968; DuBuy et al., 1973), which is associated with enhanced Fc and complement receptor activity of macrophages (Lussenhop et al., 1982).
During the first few days of infection, there is necrosis of the thymus-dependent regions of lymphoid tissues throughout the body (Hanna et al., 1970a,b; Profitt et al., 1972) and lymphocytopenia (Riley, 1968). Thymus weight decreases by 40% but returns to normal or above normal weight by seven days after infection. These changes in lymphoid tissues are abrogated by adrenalectomy prior to LDV infection (Profitt and Congdon, 1970; Santisteban et al., 1972; Brinton, 1982).
In the first few weeks of infection there is depression of cellular immunity, as evidenced by the longer survival of skin allografts (Howard et al., 1969) and increased tumor growth (Michaelides and Schlesinger, 1974); these functions gradually return to normal after weeks or months.
Antigenic challenge by T-cell-dependent antigens within 24 hours of LDV infection leads to enhanced humoral responses, while challenge three weeks or longer after infection leads to diminished responses (Mergenhagen et al., 1967; Riley et al., 1975; Isakov et al., 1982c). A similar enhanced response during early infection has been reported for a T-cell-independent antigen, but diminution of the response in chronic infection did not occur, suggesting defective T-cell function (Michaelides and Simms, 1980). Mice infected with LDV produce high blood levels of anti-LDV IgG antibodies that remain high for a year or longer (Coutelier et al., 1986).
Circulating antigen-antibody complexes are produced by four weeks post
infection. These complexes partially neutralize the virus, but it is still infectious. Complexes are deposited in glomeruli but produce only a mild membranous glomerulopathy. Protective antibody is not produced (Notkins et al., 1968; Porter and Porter, 1971).
LDV infection causes overt disease in mice of the C58 and AKR strains when they are immunosuppressed, either naturally during the aging process or experimentally by the administration of cyclophosphamide or x-irradiation. Mice of the C58 strain are more susceptible than those of the AKR strain. Beginning at five months of age C58 mice lose Lyt-1,2 cells, and the process reaches completion around 1 year of age, rendering them susceptible to age-dependent polioencephalomyelitis upon infection with LDV. Motor neurons in the anterior horn of the spinal cord are infected with LDV (Contag et al., 1986), resulting in neuronal destruction, mononuclear infiltration, and microglial proliferation in the gray matter of the cord. Clinically, there is flaccid hind limb paralysis. Polygenic inheritance of susceptibility, possibly involving the H-2 complex, has been proposed (Duffey et al., 1976; Martinez, 1979; Martinez et al., 1980; Nawrocki et al., 1980; Bentley et al., 1983). When other mouse strains are given cyclophosphamide prior to inoculation with LDV, a wide range of different nervous system lesions is produced (Stroop and Brinton, 1983).
Diagnosis of LDV infection usually is based on the finding of increased levels of LDH activity in the plasma of mice. Screening of transplantable tumors, virus inocula, and other preparations for LDV contamination is done by injecting an aliquot into LDV-free mice and performing the LDH assay on plasma or serum 72-96 hours later (Notkins, 1965; Collins and Parker, 1972; Brinton, 1982). The diagnosis of LDV infection is based on the occurrence of an 8- to 11-fold increase in LDH activity in blood plasma within the period of 72-96 hours. Only pathogen-free mice should be used for this purpose as it has been reported that prior infection with mouse hepatitis can delay the increase in LDH for five days or more (Dillberger et al., 1987b). Different bleeding techniques also can affect plasma LDH activity levels (Dillberger et al., 1987a). Plasma may be stored at room temperature for 24 hours without affecting activity (Dillberger et al., 1987a). LDV can be propagated in primary cultures of mouse tissues, but virus isolation is not practical for most diagnostic purposes (Brinton, 1982).
LDV can be eliminated from tumors by passage of tumor cells in a rodent species other than the mouse or by maintenance of tumor cells in tissue culture (Plagemann and Swim, 1963, 1966; Notkins, 1965). LDV can
be eliminated from stocks of murine plasmodia by separating parasitized erythrocytes from the mononuclear leukocytes that serve as host cells for LDV and vigorously washing the parasitized erythrocytes before passaging them in pathogen-free mice (Parke et al., 1986).
LDV-free animals can be derived from known contaminated stocks by selection of animals with normal plasma LDH concentration or by cesarean derivation. The risk of vertical and horizontal transmission from an infected dam is far greater during early stages (week one) of the infection (Notkins, 1965).
LDV infection is not likely to be encountered in mice from commercial barrier breeding facilities.
Interference with Research
Riley (1974) has aptly described LDV as "the benign modifier of body chemistry." Although LDV affects mainly the immune system and concentrations of certain enzymes in the plasma, its potential for altering research data is enormous and complex. It is extremely subtle because infection is subclinical throughout life. Its effects on many biologic endpoints can differ dramatically with time after infection (e.g., one week after infection versus weeks or months after infection). In addition, subtle interactions with other agents can occur (e.g., immunosuppression caused by LDV can alter defenses against other infectious agents).
LDV infection causes an 8- to 11-fold increase in plasma LDH and a two- to three-fold increase in several other plasma enzymes (Brinton, 1982).
LDV infection can enhance or suppress growth of transplantable mouse tumors. In general, tumor growth is enhanced early after LDV infection (because of depressed cellular immunity) and is influenced less during chronic infection. Some examples include Ehrlich ascites tumor (Bailey et al., 1965a), MOPC-315 plasmacytoma (Michaelides and Schlesinger, 1974), Gardner lymphoma (Speckman and Riley, 1975), and a chemically induced fibrosarcoma (Henderson et al., 1979).
LDV infection results in the altered incidence and behavior of spontaneous virus-induced neoplasms, including the Bittner mammary tumor (Riley, 1966) and murine sarcoma virus (Turner et al., 1971; McDonald, 1983). LDV infection has been reported to suppress the development of pulmonary adenomas in response to urethan (Theiss et al., 1980) and carcinogenesis caused by vinyl chloride-vinyl acetate (Brinton and Brand, 1977).
LDV infection does not alter the development of spontaneous reticulum cell sarcoma and lymphatic leukemia in SJL/J, AKR/Cu, and AKR/J mice, and does not influence carcinogenicity of methylcholanthrene or benzanthracene (Isakov et al., 1981).
LDV infection causes delayed allograft rejection in mice (Howard et al.,
1969), prevents development of experimental allergic encephalomyelitis in mice (Inada and Mims, 1986), and prevents the occurrence of autoimmune disease in NZB and (NZB x NZW)F1 mice (Oldstone and Dixon, 1972).
LDV infection causes elevation of serum gamma globulin levels and increased humoral antibody responses during early infection (Notkins et al., 1966; Mergenhagen et al., 1967; Riley et al., 1975; Michaelides and Simms, 1980; Isakov et al., 1982c) and a reduced humoral response during chronic infection (Michaelides and Simms, 1980).
LDV is a polyclonal lymphocyte activator during the early stages of infection (Michaelides and Simms, 1980).
LDV inhibits carbon clearance (Notkins and Scheele, 1964; Mahy et al., 1965) and reduces the plasma clearance of injected asparaginase (Riley et al., 1970).
LDV infection greatly enhances the severity of Eperythrozoon coccoides infection; the dual infection results in severe hemolytic anemia (Riley, 1964; Fitzmaurice et al., 1974). It exacerbates murine malaria caused by Plasmodium yoelii (Henderson et al., 1978) and increases susceptibility to experimental Listeria monocytogenes infection (Bonventre et al., 1980).
An Ehrlich ascites tumor was found to induce interferon and enhance natural killer cell activity when injected into mice, and both effects were traced to LDV contamination (Koi et al., 1981). Interferon induction occurred after injection of a mouse monoclonal antibody into mice, and the cause was shown to be LDV contamination (Nicklas et al., 1988).
Haemobartonella muris has little significance for most experimental uses of rats. It has high significance for studies involving rat-to-rat passage of materials (e.g., transplantable tumors or inocula for experimental blood parasite infections).
Since Mayer (1921) discovered this agent as a complicating factor in studies of experimental Trypanosoma brucei infection in rats, it has occasionally been rediscovered by unwary experimentalists. In 1960 Sacks and co-workers (Sacks and Egdahl, 1960; Sacks et al., 1960) rediscovered it as a "filterable hemolytic anemia agent," and it was correctly identified later by Moore et al. (1965). Thus, this usually subclinical infection of rats and contaminant of biologic materials from rats can be a subtle complication of research.
Haemobartonella muris is a bacterium, order Rickettsiales, family Anaplasmataceae. It is Gram negative; cocci (100-500 nm in diameter), diplococci, or slender rods (100 nm in diameter x 300-700 nm in length); obligately parasitic and occurs in indentations on the erythrocyte surface or in vacuoles within erythrocytes and, rarely, free in plasma. It is not cultivable outside the host. Growth in the host is inhibited by arsenicals and tetracyclines (Ristic and Kreier, 1984).
Viability is rapidly lost after 0.5 hour at 37°C, 6-8 hours at 25°C, and 24-48 hours at 4°C. It can be preserved indefinitely in 10% dimethyl sulfoxide stored in liquid nitrogen (Ristic and Kreier, 1984).
Rats. Possibly mice and hamsters (Ristic and Kreier, 1984).
Natural transmission is by the spined rat louse Polyplax spinulosa (Crystal, 1958, 1959a,b). Transplacental and oral transmission have been suggested (Weinman, 1944), but definitive proof of both possibilities is lacking (Crystal, 1958, 1959a,b; Owen, 1982).
Transmission is readily accomplished by injecting biologic materials (e.g., blood, transplantable tumors, tissue homogenate) contaminated with the agent via parenteral routes. Thus, inadvertent transmission during experimental procedures constitutes an important source of infection.
Infection can persist throughout life without clinical signs, unless it is activated by natural or experimental immunosuppression.
During active disease, signs can include anemia, pallor, dyspnea, weight loss, and hemoglobinuria. A hemogram can reveal anemia, reticulocytosis, increased IgG and IgM, reduced plasma proteins, increased clotting time, and terminal hyperphosphatemia (Kessler, 1943; Kessler and Zwemer, 1944; Baker et al., 1971; Finch and Jonas, 1973; Cox and Calaf-Iturri, 1976; Lindsey et al., 1978b).
In natural infections there are no lesions, and parasitemia cannot usually be detected by examination of stained blood smears. Active disease is char-
acterized by anemia, hemoglobinuria, splenomegaly, and parasitemia (Baker et al., 1971).
Activation of infection by splenectomy followed by demonstration of parasitemia can be used to diagnose the infection in individual animals. Alternatively, rats known to be free of Haemobartonella muris and other pathogens can be splenectomized and then injected parenterally with transplantable tumor homogenates, pooled blood from groups of rats, or other test materials, followed by attempts to demonstrate parasitemia (Baker et al., 1971; Cassell et al., 1979).
Animals that are several months of age are more susceptible than young rats to severe disease following splenectomy. After splenectomy or injection of the test material into previously splenectomized rats, parasitemia usually appears in 2-6 days, the erythrocyte count may drop to less than one million/ mm3, and death may ensue because of a hemolytic crisis (Baker et al., 1971; Cassell et al., 1979).
In blood smears stained by the Giemsa or Romanowsky methods, organisms appear as coccoid, dumbbell, or long rod forms on erythrocytes. Transmission electron microscopy can be used for confirmation (Tanaka et al., 1965). Organisms must be differentiated from the basophilic stippling that is common in rodent erythrocytes (Griesemer, 1958; Baker et al., 1971; Cassell et al., 1979).
Cesarean derivation and barrier maintenance apparently have been very successful in eliminating the infection from breeding colonies. Polyplax splinulosa must be controlled.
Rat tumors, cells, and other biologic materials to be passaged in rats should be screened to ensure the absence of the agent (Baker et al., 1971). Organic arsenicals and tetracyclines are reported to eliminate the organism from hosts with either latent or active infection (Ristic and Kreier, 1984).
Interference with Research
Because of their usual clinically silent character, H. muris infections may be extremely subtle causes of variability in certain biologic responses. Infection has been shown to do the following:
- Reduce the half-life of erythrocytes (Rudnick and Hollingsworth, 1959).
- Modulate the course of experimental malaria (Hsu and Geiman, 1952) and trypanosomiasis (Marmorston-Gottesman and Perla, 1930).
- Enhance phagocytic activity, e.g., clearance of intravenously injected carbon (Elko and Cantrell, 1968).
- Increase the rejection rate of transplantable tumors (Sacks et al., 1960).
- Cause fulminant hemolytic anemia in recipient rats given H. muris-contaminated transplantable tumors (Sacks and Egdahl, 1960; Sacks et al., 1960).
Subclinical H. muris infections are activated by the ablation of splenic phagocytes by methods including splenectomy (McCluskie and Niven, 1955; Scheff et al., 1956), injection of anti-rat spleen serum (Pomerat et al., 1947; Thomas et al., 1949), intravenous injection of ethyl palmitate (Stuart, 1960; Finch and Jonas, 1973), injection of Polonium-210 (Scott and Stannard, 1954), and whole body x-irradiation (Rekers, 1951; Scott and Stannard, 1954; Berger and Linkenheimer, 1962). Also, latent infections can be activated by experimental infections of Plasmodium sp. (Hsu and Geiman, 1952).
Eperythrozoon coccoides has little significance for most experimental uses of mice. This agent is highly significant for studies involving mouse-to-mouse passage of materials such as transplantable tumors or inocula for experimental blood parasite infections.
This is an agent that is subject to occasional rediscovery by investigators who passage biologic materials in mice without monitoring for contaminating pathogens. A classic example is the work of Stansly and co-workers (Stansly et al., 1962; Ansari et al., 1963), who were attempting to demonstrate an oncogenic virus by passaging in mice splenic filtrates from leukemic mice. They described a filterable agent and named it the spleen weight increase factor or SWIF, only to learn later that they had rediscovered Eperythrozoon coccoides (Stansly and Neilson, 1965).
Demonstration that E. coccoides infection enhances the disease caused by mouse hepatitis virus infection in mice and separation of these two indigenous agents by Gledhill and associates (Gledhill and Andrewes, 1951; Niven et al., 1952; Gledhill and Dick, 1955: Gledhill et al., 1955: Gledhill, 1961) are important historical landmarks in the field of laboratory animal disease.
Eperythrozoon coccoides is a bacterium, order Rickettsiales, family Anaplasmataceae. It is a Gram-negative, coccoid organism, measuring 350500 nm in diameter. It is obligately parasitic, and occurs loosely attached to erythrocytes and free in blood plasma. It grows in embryonated hen's eggs. Growth in the host is inhibited by neoarsphenamine and tetracyclines (Kreier and Ristic, 1968; Ristic and Kreier, 1984).
E. coccoides readily passes coarse bacterial filters. It has an unusually low specific gravity, and is not completely sedimented by centrifugation of infected plasma at 100,000 x g for 1 hour. It is inactivated rapidly by disinfectants and drying. Freezing (-70°C) preserves its infectivity, while 37°C for 3 hours destroys its infectivity (Stansly and Neilson, 1966; Baker et al., 1971).
Mice. Possibly rats, hamsters, and rabbits (Ristic and Kreier, 1984).
Polyplax serrata, the mouse louse, is considered the main vector; it transmits the infection mechanically (Berkenkamp and Wescott, 1988). Transmission by oral, intranasal, and transplacental routes has been suspected but not proved. The infection persists throughout life (Baker et al., 1971).
Reliable data on prevalence of E. coccoides in contemporary mouse stocks are lacking. The agent is most likely to be encountered in conventionally reared stocks of mice that have not been cesarean derived and maintained by barrier methods (Baker et al., 1971).
Inadvertent transmission in biologic materials (such as transplantable tumors, inocula for passage of experimental parasitic infections, blood plasma, and cell-free filtrates) rather than natural transmission poses the greatest threat in modern research (Baker et al., 1971; Lindsey et al., 1978b; Iralu and Ganong, 1983).
Natural and experimental infections of E. coccoides usually are inapparent.
The spleen weight increases to three-four times normal by post infection day seven, reduces to one and one-half to two times normal by day 21, and persists at this latter size at least until day 42. Microscopically, there are
increased erythroid elements in the enlarged spleen and increased fixed macrophages in the spleen and liver (Thurston, 1955; Baker et al., 1971; Cox and Calaf-Iturri, 1976; Lindsey et al., 1978b).
Detection of subclinical E. coccoides infection is usually by splenectomy to produce active infection, followed by demonstration of the agent in peripheral blood. Following splenectomy, persistently infected mice can develop massive parasitemia within 2-4 days, but it can be surprisingly short-lived (a few to 24 hours). Thus, detection of the organism may require frequent examination of blood (e.g., every 6 hours). The preferred approach is to splenectomize several pathogen-free mice. These mice are then inoculated intraperitoneally or intravenously with test material that is to be monitored for the infection (e.g., pooled blood or spleen from mice of a colony suspected of having latent infection, transplantable tumors, antisera, cell lines). Peripheral blood is examined for the organism every 6 hours for four days. The organism can be demonstrated in peripheral blood by light microscopic examination of stained blood smears, electron microscopy, acridine orange, and indirect fluorescent antibody (IFA) test. In Giemsa- or Romanowsky-stained blood smears, organisms appear as distinctive ring forms on erythrocytes and free in plasma. IFA is particularly useful for detecting small numbers of organisms (Berkenkamp and Wescott, 1988). The organism must be differentiated from the basophilic stippling that is common in rodents (Tanaka et al., 1965; Baker et al., 1971).
Cesarean derivation followed by barrier maintenance (including ectoparasite control) are thought to be effective, but definitive data are lacking. Treatment with neoarsphenamine or tetracycline can be useful in selected situations. The most important approach is to prescreen tumors and other biologic materials to be passaged in mice to ensure the absence of the agent (Baker et al., 1971).
Interference with Research
E. coccoides infection has been shown to cause the following:
- Suppress interferon production in mice during the first 3 weeks after infection (Glasgow et al., 1971, 1974).
- Exacerbate concurrent infections of mouse hepatitis virus (Niven et al., 1952; Nelson, 1953; Gledhill and Dick, 1955; Gledhill et al., 1955; Gledhill and Niven, 1957), lymphocytic choriomeningitis virus (Seamer et
- al., 1961), or lactate dehydrogenase-elevating virus (Arison et al., 1963; Riley, 1964; Riley et al., 1964).
- Reduce pathogenicity of concurrent infection by Plasmodium sp. (Peters, 1965; Ott and Stauber, 1967).
- Increase phagocytic activity in mice (Gledhill et al., 1965).
- Increase susceptibility of mice to bacterial endotoxins (Gledhill and Niven, 1957).
The agent can be a common contaminant of inocula used to perpetuate experimental blood parasite infections in mice (Peters, 1965; Ott and Stauber, 1967; Iralu and Ganong, 1983).
Subclinical infection can be activated by whole body x-irradiation (Berger and Linkenheimer, 1962) or splenectomy (Marmorston, 1935).
Murine Leukemia Viruses
This large group of genetically related viruses, referred to collectively as the murine leukemia viruses (MuLVs), has provided a wide range of model systems for studies of the molecular biology, virology, genetics, pathology, immunopathology, and experimental chemotherapy of leukemias.
The discovery of the MuLVs was linked historically to the development of inbred strains of mice, many of which were selected for their susceptibility or resistance to MuLV (Heston, 1974, 1975). Furth et al. (1933) developed the mouse strain AKR, which has a high incidence of leukemia. Furth (1946) later showed that thymectomy prevented the leukemia in these mice. In 1951, Gross (1951a,b) demonstrated that when cell-free filtrates of leukemic tissues from a strain of mouse now known as AKR were injected into infant C3H/Bi mice, leukemia was successfully transmitted. This led to the discovery of numerous strains of MuLVs in transplantable murine tumors, including those described by Graffi (1957), Friend (1957), Schoolman et al. (1957), Moloney (1960), Rauscher (1962), and Tennant (1962). Gross (1959) and Lieberman and Kaplan (1959) demonstrated that radiation-induced lymphoid tumors of mice also contained MuLVs. Subsequently, MuLVs were studied extensively by using both in vivo and in vitro systems, which contributed many of the modern concepts of viral oncology, such as the oncogene theory (Huebner and Todaro, 1969) and the role of reverse transcriptase (Baltimore, 1970). In no instance, however, have the basic mechanisms of leukemogenesis been fully explained (Furmanski and Rich, 1982).
The MuLVs are RNA viruses, family Retroviridae, subfamily Oncovirinae, genus type C oncovirus group, subgenus mammalian type C oncovirus, species murine leukemia and sarcoma viruses. Related viruses infecting mammalian hosts include baboon type C oncovirus; bovine leukosis virus; feline sarcoma and leukemia viruses; gibbon ape leukemia virus; guinea pig type C oncovirus; porcine type C oncovirus; rat type C oncovirus; woolly monkey sarcoma virus; and, more recently, the human T-cell lymphotropic virus (HTLV), currently referred to as the human immunodeficiency virus (HIV) (Lieber and Todaro, 1975; Poiesz et al., 1981; Matthews, 1982).
Like other members of the Retroviridae, MuLVs have a spherical, enveloped virion measuring 80-100 nm in diameter with glycoprotein surface projections of approximately 8 nm in diameter. The genome is a dimer of linear, positive-sense, single-stranded RNA. Genetic information for production of infectious progeny consists of three genes: gag, which codes for nonglycosylated virion proteins; pol, which codes for reverse transcriptase; and env, which codes for envelope glycoproteins. The env glycoproteins are responsible for type specificity, while the major internal gag protein defines group (subgenus) specificity (Matthews, 1982).
Numerous strains of MuLV have been recognized. Strains usually are designated by their discoverer's name(s) or by symbols assigned by the original investigators. Different strains of MuLVs have 80% genetic homology and are very prone to undergo mutation or recombination. They are generally classified on the basis of their host range (infectivity) for tissue culture cells. The host range is probably determined by the env gene product, gp70 protein. The ecotropic (or mouse tropic) MuLVs replicate only in mouse cells. However, the replication of MuLVs in cells of different mouse strains is restricted by the Fv-1 locus. The xenotropic strains grow only on nonmurine cells, e.g., mink, rabbit, human, and duck. Amphotropic strains replicate in both mouse and nonmurine cells. Polytropic [or dualtropic or mink cell focus (MCF)] strains grow on mouse and nonmurine cells but differ from amphotropic strains by serologic neutralization. The amphotropic strains occur mainly in wild mice. The dualtropic strains are laboratory-derived recombinant viruses (Steeves, 1974; Hartley et al., 1977; Famulari, 1983; Risser et al., 1983; Goff, 1984).
Laboratory and wild Mus musculus are considered the natural hosts. Experimentally, leukemias have been induced by the inoculation of MuLV into neonatal mice, rats, and hamsters (Steeves, 1974).
There are two distinct phases in the life cycle of the type C oncoviruses. In one phase, the virus genome exists as DNA integrated in the host cell genome (the so-called provirus), where it is replicated and transmitted to all daughter cells along with the host cell genome. In the second phase, complementary RNA sequences are synthesized from the DNA provirus and packaged in extracellular virions for transmission to other host cells. In the target cells there is reverse transcription of the RNA genome into DNA and integration of the DNA into the host cell genome (Bishop and Varmus, 1975; Lilly and Mayer, 1980).
The endogenous type C oncoviruses are integrated in the DNA of the host's sex cells and are transmitted vertically as Mendelian traits. All laboratory and wild M. musculus are thought to harbor endogenous type C oncoviruses. Horizontal transmission is inefficient but can occur by transfer of virus in saliva, sputum, urine, feces, or milk or by intrauterine infection. The exogenous type C oncoviruses, of which the Friend, Moloney, and Rauscher strains are prototypes, are laboratory variants that have been derived from transplantable tumors and are known to become integrated only in the DNA of somatic cell lines (Lieber and Todaro, 1975; Lilly and Mayer, 1980).
Despite the fact that all mice have endogenous MuLVs, leukemias and related malignancies occur naturally in only 1-2% of most strains (Furmanski and Rich, 1982). In contrast, AKR mice have a high incidence of spontaneous thymic lymphoma, reaching 90% by 9 months of age. In affected mice, the most common signs are dyspnea (due to thymic lymphoma), peripheral lymphadenopathy, or abdominal enlargement (Squire et al., 1978).
The mechanisms of MuLV expression (replication) and leukemogenesis are extremely complex. The MuLVs are a collection of viruses that are related in genetic sequence but that have diverse patterns of host range and tissue tropism. All mouse strains have MuLV genomes (proviruses) for ecotropic and xenotropic viruses at different loci in their genomes. Mouse strains with a low incidence of leukemia (BALB/c, A/J, C3H/He, and CBA/ J) have single ecotropic proviruses at a few nonallelic chromosomal loci, while mouse strains with a high incidence of leukemia (AKR, C58, C3H/ Fa) have multiple ecotropic proviruses at nonallelic chromosomal loci. The
chromosome number and precise location of these loci is strain dependent. Mouse strains with a high incidence of leukemia spontaneously express high titers of ecotropic MuLV in all organs early in life, and these persist throughout life. In contrast, mouse strains with a low incidence of leukemia express only low titers of virus. Expression of the viral genomes in mice with a low incidence of leukemia, however, can be induced by chemical carcinogens, radiation, and other stimuli. The complex mechanisms by which endogenous MuLVs induce leukemia, in mouse strains with either a low or high incidence of leukemia, appear to involve the expression and interaction of multiple MuLV genomes of both ecotropic and xenotropic viruses through such processes as recombination, DNA transfection, and transcomplementation (Furmanski and Rich, 1982; Morse and Hartley, 1982; Famulari, 1983; Risser et al., 1983; Goff, 1984). Host susceptibility is influenced by a number of mouse genes including those of the Fv-1, Fv-2, In, nu, and hr loci and the Ir locus of the major histocompatibility complex (Lilly and Pincus, 1973; Heiniger et al., 1974; Steeves and Lilly, 1977; Lilly and Mayer, 1980; McCubrey and Risser, 1982; Meruelo and Bach, 1983).
The mouse leukemias are, with few exceptions, actually lymphomas because they are predominantly solid tumors of lymphocytes or other hematopoietic cells. The spontaneous tumors in mice are mainly of two types. The majority of those occurring before 1 year of age are thymic lymphomas, while the predominant type seen in older mice is histiocytic (reticulum cell) lymphomas (Dunn, 1954; Squire et al., 1978; Della Porta et al., 1979).
The classification scheme and morphologic descriptions of the spontaneous and experimentally induced hemopoietic neoplasms reported by Dunn (1954), with minor modification, serve as the standard for diagnosis. This family of neoplasms includes thymic lymphomas, nonthymic lymphomas, histiocytic (reticulum cell) lymphomas, lymphatic leukemias, granulocytic leukemias, erythroleukemia, plasma cell tumors, and mast cell tumors (Dunn, 1954; Squire et al., 1978; Della Porta et al., 1979).
The incidence of spontaneous hemopoietic neoplasms occurring in mice has been the subject of numerous investigations. For comparative data and references on many strains of mice, the reader is referred to the tabulated information given by Squire et al. (1978) and Furmanski and Rich (1982).
The diagnosis of spontaneously occurring hemopoietic neoplasms in the mouse is usually made on the basis of morphologic features in histologic sections (Dunn, 1954; Squire et al., 1978; Della Porta et al., 1979). Tumor cell types also can be determined by analysis of cell surface antigens (Old
and Stockert, 1977). The isolation and characterization of MuLVs require specialized techniques usually available only in research laboratories dedicated to viral oncology. However, MuLV type and group specificity can be determined by immunofluorescence of fixed cells by use of an appropriate panel of type- or group-specific antisera (Matthews, 1982).
All mice probably have vertically transmitted endogenous MuLV proviruses as an integral part of their genomes; this is true even for mice that are otherwise considered germfree (Kajima and Pollard, 1965). Horizontal transmission can occur but is considered inefficient (Furmanski and Rich, 1982). Therefore, control measures are not usually considered useful unless mice are being infected experimentally with high doses of MuLV, in which case the infected mice should be isolated from uninfected control mice.
Interference with Research
Although all mice have endogenous MuLVs, their presence probably has little significance for most research purposes in which mice are used. MuLV expression and the associated occurrence of neoplasms, however, can present competing endpoints in some studies, e.g., studies of the aging processes in various organs. Thus, an awareness of spontaneous tumors in different strains of mice or the induction of tumors by specific test chemicals can be useful in the design of some experiments. Also, active MuLV infection can cause suppression of humoral and cellular immunity without clinically obvious disease (Specter et al., 1978).