key: cord-0883765-lno01041
authors: Buchmeier, Michael J.; Lewicki, Hanna A.; Talbot, Pierre J.; Knobler, Robert L.
title: Murine hepatitis virus-4 (strain JHM)-induced neurologic disease is modulated in Vivo by monoclonal antibody
date: 1984-01-30
journal: Virology
DOI: 10.1016/0042-6822(84)90033-3
sha: db9141472d70bd95d435fa5dd2245682d627e051
doc_id: 883765
cord_uid: lno01041

Abstract Monoclonal hybridoma antibodies directed against the polypeptides of murine hepatitis virus-4 (JHM strain) were tested for their ability to alter the course of a normally lethal intracerebral virus challenge. Three monoclonal antibodies directed against two distinct epitopes on the E2 glycoprotein of MHV-4 protected mice against lethal virus challenge and converted the infection from fatal encephalomyelitis to demyelination. A single neutralizing antibody directed against a third epitope on E2 as well as seven nonneutralizing antibodies to E2, E1, and N polypeptides did not protect against challenge. In mice which received protective antibody, MHV-4 infection was not blocked, however, virus grew to lower titers in liver and brain, and virus replication in the CNS was more restricted than in unprotected mice. Decreased involvement of neurons in the brains of protected mice was observed, and no evidence of neuronal infection in the spinal cords was found. In contrast, oligodendrocytes were infected in the presence of protective antibody, and evidence of demylination associated with mononuclear cell infiltration was found. These studies demonstrate that antibody to a single epitope on a viral glycoprotein can substantially alter the course and phenotype of disease.

hybridoma antibodies directed against the polypeptides of murine hepatitis virus-4

(JHM strain) were tested for their ability to alter the course of a normally lethal intracerebral virus challenge. Three monoclonal antibodies directed against two distinct epitopes on the E2 glycoprotein of MHV-4 protected mice against lethal virus challenge and converted the infection from fatal encephalomyelitis to demyelination. A single neutralizing antihody directed against a third epitope on E2 as well as seven nonneutralizing antibodies to EZ, El, and N polypeptides did not protect against challenge. In mice which received protective antibody, MHV-4 infection was not blocked, however, virus grew to lower titers in liver and brain, and virus replication in the CNS was more restricted than in unprotected mice. Decreased involvement of neurons in the brains of protected mice was observed, and no evidence of neuronal infection in the spinal cords was found. In contrast, oligodendrocytes were infected in the presence of protective antibody, and evidence of demylination associated with mononuclear cell infiltration was found. These studies demonstrate that antibody to a single epitope on a viral glycoprotein can substantially alter the course and phenotype of disease.

Murine hepatitis virus-4 (MHV-4) strain JHM is a neurotropic member of the coronaviridae.

Infection by MHV-4 in the mouse is associated with encephalitis and demyelination (Bailey et al, 1949; Waksman and Adams, 1962; Weiner, 1973) . Fatal encephalitis associated with significant loss of neurons is the normal outcome of intracerebral inoculation, with resistance of mice to fatal disease apparently being controlled by a single autosomal recessive gene (Knobler et al, 1981a) expressed at the level of the neuron and macrophage. Recovery from the acute encephalitis is rare, and the infrequent survivors exhibit demyelination (Lampert et d, 1973; Weiner, 1973) . Haspel and co-workers (1978) a temperature sensitive mutant of MHV-4, designated ts 8, produced demyelination in over 90% of infected mice while producing encephalitis in less than 5%. The ts 8 mutant infects oligodendrocytes leading to cell death, degeneration, and removal of their myelin sheaths by macrophages (Knobler et d, 1982) .

Mouse hepatitis viruses contain at least three major classes of structural proteins including the 50,000-Da nucleocapsid protein (N), and two glycoproteins El and E2 (Wege et aZ., 1979; Sturman et aL, 1980; Holmes et al, 1981) . El is integrally associated with the membrane in both a nonglycosylated Z&000-Da form and an O-glycosylated 25,000-Da form. E2 consists of two approximately 95,000-Da subunits which form a 180,000-Da dimer constituting the large "petal" (peplomer) on the virion envelope (Sturman et al, 1980) . In our laboratory we have raised a library of hybridoma antibodies to these viral proteins to study the biology and biochemistry of MHV and to probe the in vivo events in VimLs neutralization. Virus neutralizing disease. With these antibodies we were capacity of ascites fluids containing monoable, in our initial studies, to assign the clonal antibody was quantitated by a viral components responsible for attach-plaque reduction neutralization assay. Asment and cell-cell fusion to the E2 gly-cites fluids were diluted in MEM as indicoprotein (Collins et ak, 1982) , and to map cated and mixed with an equal volume (0.5 the epitopic regions of these molecules ml) of virus dilution containing 120 to 200 (Talbot et aL, 1984) . In this report we have PFU of MHV-4. The virus-antibody mixstudied the ability of antibodies to specific tures were incubated at 37" for 30 min, viral proteins to alter the course of disease divided in half, and plated in duplicate on in vivo, and have mapped the property of monolayers of L-24 cells in 60-mm culture passive protection against lethal enceph-dishes. After adsorption for 1 hr, overlay alitis to specific epitopes on the E2 gly-medium was added and the plates were coprotein.

further incubated 72 hr in a 37" COz incubator. Cells were fixed by the addition of 2 ml of 25% formalin in PBS for 4-18 MATERIALS AND METHODS hr. Agar overlays were removed and monolayers stained with 0.1% crystal vi-Virus and cell culture. MHV-4 (JHM olet. Neutralization was expressed as the strain) was originally obtained from Dr. reciprocal of the antibody dilution giving Leslie Weiner and is routinely propagated 50% reduction in plaque number compared on L-24 cells as previously described (Has-with virus incubated in parallel with culpel et al, 1978; Collins et al., 1982) . The ts ture medium not containing antibody. As-8 mutant MHV-4, which produces a high cites fluids directed against nonneutralfrequency of demyelinating disease with-izing antigens of MHV-4 (nucleocapsid out encephalitis, was isolated by Haspel et protein) or against other unrelated viruses c& (1978) . Virus was enumerated by plaque often exhibited low levels of an MHV-4 assay on L-24 cells.

neutralizing activity (~300 PRD,/ml); Hybridwmu cells. Hybridoma cells pro-thus a control sample consisting of nuducing monoclonal antibodies to MHV-4 cleocapsid antibody was included in all aswere generated and characterized in this says. laboratory as previously described (Collins Passive antibody protection and assesset &, 1982) . All lines were cloned by lim-ment of demyelination BALB/c St mice iting dilution in 96-well plates at a ratio aged 4-6 weeks are exquisitely sensitive to of 0.1 cell per well. Specificity for viral intracerebral (ic) infection with MHV-4 polypeptides and epitopes was established (Knobler et al, 1981a) , and develop a rapas described elsewhere (Collins et al, 1982 ; idly fatal encephalitis due to infection of Talbot et al, 1984) .

neurons. We have used such ic infected Ascites fluids were produced in BALB/ mice as a test system to assess the proc St mice. Briefly, mice were pretreated by tective effect of passively administered ip injection with 1 ml of Pristane (2, monoclonal antibodies on MHV-4 infection. 6,10,14-tetramethylpentadecane, Aldrich Two protocols of antibody administration Chemical Co., Milwaukee, Wise.). Seven days later, 5 X lo6 to 1 X 10' hybridoma were employed. In initial experiments, we gave mice daily doses of 25 ~1 of undiluted cells were injected intraperitoneally.

Asascites on Days -2, -1, 1,2,3, and 4 relative cites developed lo-14 days later, and were to virus challenge on Day 0. In order to harvested daily from unanesthetized mice minimize handling of mice after infection, by puncture with a 20-gauge needle. Ascites we subsequently adopted a protocol of a fluids were clarified by centrifugation at 2500 g for 20 min, aliquoted, and stored at single dose of 200 ~1 of ascites on Day -1 and found that this regimen was as efficient -20". Immunoglobulin concentrations in as the first at conferring protection. In all ascites fluids were estimated by radial imcases, virus challenge consisted of lo-50 munodiffusion.

PFU (approximately 30-150 LD& of wild-type MHV-4 intracerebrally. This dose reproducibly yielded 100% mortality by 6 days after infection.

To assess demyelination in passively protected mice, we allowed survivors to live for 14-21 days after virus challenge, then sacrificed and perfused them via the left ventricle with glutaraldehyde-paraformaldehyde fixative in PBS as previously described (Knobler et al. (1981b) . Following perfusion, spinal cords and brains were dissected out and l-pm Epon sections were prepared and stained for myelin with-p phenylenediamine.

Surveys for demyelinating foci routinely included examination of two sections each from cervical and lumbar spinal cord and one section of thoracic cord. An animal was judged demyelinated if foci of axonal demyelination were observed in the white matter of any of these five sections. Immunoperoxidase staining for viral antigen was performed on 30-pm vibratome sections as previously described (Knobler et aL, 1981b) using protein A peroxidase and monospecific rabbit antibody to MHV-4 (kindly supplied by K. V. Holmes, USPHS, Bethesda, Md.). For antigen studies, glutaraldehyde was omitted from the perfusion fixative.

For virus titration, 10% homogenates of specified tissues were prepared from mice infected and treated as indicated and assayed on monolayers of L-24 cells. Virus titers were expressed as PFU per gram of tissue.

bg mcmoclonal antibodies. Previous work from our laboratory established that MAb to the MHV-4 glycoprotein E2 neutralized virus in vitro (Collins et d, 1982) . We have extended these findings in surveying a larger panel of antibodies to the El and E2 glycoproteins using a more sensitive plaque reduction assay in which we determined PRD, titers for the monoclonal ascites preparations used in the present studies. Titers observed ranged from approximately 8000 PRD,/ml to greater than 158,000 PRD,/ml for various ascites preparations as summarized in Table 1. Note that a low background titer generally less than 300 PRDdml was observed in preparations of nucleocapsid specific antibody as well as in unrelated monoclonal ascites and in some normal mouse sera (data not shown). Whether this naturally occurring activity represents specific antibody or a cross-reacting natural antibody is under investigation.

Passively transferred MAb blocks lethal encephalitis induced by Intracerebrally inoculated BALB/c mice are exquisitely sensitive to MHV-4. Infection of neurons of the CNS results in rapidly lethal encephalitis usually within 6 days. We tested two passive transfer regimens as described under Materials and Methods to attempt to alter the course of this normally lethal encephalitis. Both regimens were effective, and the simpler single antibody dose protocol was adopted. Figure 1 shows the results of one such protection experiment in which we tested five different MAb to the MHV-4 E2 glycoprotein (5B19.2, 5A13.5,5B170.3,4B11.6, and 5B93.9) and as a control, a single antibody to the N protein (4B6.2). Groups of six mice were given 200 ~1 of the indicated ascites on Day-l then challenged intracerebrally with 20 PFU of MHV-4 on Day 0. Survivors were observed daily for 21 days then sacrificed in order to score demyelinating lesions in the spinal cord. Note that mice receiving nonprotective antibody (clones 4B11.6, 5B93.9, and Timing and dose dependence of passive protecticm, The effect of antibody given passively prior to infection might be argued to be simply an in vivo neutralization whereby incoming virus in the challenge inoculum is prevented from infecting targets in the CNS. To test this possibility, we altered the time of antibody transfer relative to challenge. Groups of six mice each were given a single 200-~1 dose of MAb 5B19.2 on Days -1, 0, +l, and +2 relative to virus challenge on Day 0. Mortality was again scored over a 21-day interval. As evident in Fig. 2 , groups of six mice each receiving no antibody or MAb 4B6.2 on Day -1 all died by 6 days after infection as previously observed. In contrast, mice receiving antibody on Days -1, 0, and +l showed 100, 100, and 83% survival, respectively, while 33% of those receiving antibody 2 days after infection survived for 21 days. Thus, the effect of passively transferred antibody is therapeutic and confers protection even if given after the establishment of CNS infection. To examine the quantity of antibody required to protect against lethal challenge, we gave groups of six mice increasing doses of 5B19.2 ascites 1 day prior to ic challenge with MHV-4 then scored mortality as indicated above. Table 2 summarizes the results of a typical experiment. A single dose of 25 ~1 of 5B19.2 ascites with a neutralizing titer of l:lO,OOO protected half of the challenged mice. This dose of ascites in a 25 g mouse is approximately equivalent to a l/ 1000 dilution (1 bl/g).

Eflect of passively transferred antibody on virus growth in tissues. The observation that MAb was protective even if given after challenge suggested that the sparing effect operated by blocking or slowing the spread of virus infection. To investigate this, we titered MHV infectivity in tissues following infection in the presence and absence of protective antibody. Brains and livers were titered to assess the effect of antibody on replication in CNS and peripheral tissues, respectively.

MAb was passively transferred 1 day before ic challenge as above and tissues were removed for titration 4 days after infection, at a time when sufficient unprotected mice remained alive for study. Table 3 summarizes these results. We noted that titers in the brains of mice receiving MAb 5B19.2 were reduced by a factor of 20 relative to untreated controls. This reduction was not observed in mice receiving nonprotective antibody to El (5A5.2) or E2 (4B11.6). In the protected mice (group IV, Table 3 ), we were unable to detect virus in the livers at a threshold value of 200 PFU per gram whereas control mice (groups I and III) had greater than 2 X lo4 PFU per gram in their livers on Day 4. To confirm that this difference reflected actual reduction in virus replication and not merely a damping effect due to high concentration of MAb in the liver homogenates, we examined H and E sections as well as fluorescein-stained cryostat sections of livers from protected and unprotected mice. Mice which received no protective MAb showed numerous large necrotic foci of hepatocytes (Fig. 3A) , stained for MHV-4 antigens (Fig. 3B ), but only limited isolated foci of infected cells were observed in sections of livers from mice receiving protective MAb 5B19.2 (Figs. 3C, D). Taken together, these findings indicate that virus replication is substantially decreased by passively administered MAb both in the CNS and peripheral compartments. ET AL. observation of lower virus titers in the brains of protected mice suggested that virus replication was restricted to fewer cells in these animals. To test this, we compared the distribution of MHV antigens in sections of spinal cord among infected and untreated, infected and unprotected, and infected and protected mice. Figure 4 demonstrates peroxidase-labeled MHV antigens in a section from an MHV-l-infected untreated mouse. There are MHV antigens in neuronal cells in the gray matter (Fig.  4A ) as well as in oligodendrocytes with their processes surrounding multiple myelin sheaths in the white matter (Fig. 4B) .

The MHV-4 antigen distribution in infected animals given nonprotective antibody is virtually identical. In contrast, mice infected with MHV-4 and protected by antibody have a different distribution of MHV antigens. Figure 5 demonstrates peroxidase-labeled MHV antigen in oligodendrocytes, identified by their cytoplasmic processes surrounding multiple myelin sheaths (Knobler et uL, 1981b (Knobler et uL, , 1982 , in the white matter. Neurons containing MHV antigen were not usually found in the spinal cord of antibody-protected mice. These findings were corroborated by histopathology.

Both myelination. Neurons of mice infected with MHV-4 and protected by antibody were spared, but these animals did have definite areas of demyelination (Fig. 6A) . These demyelinated foci were frequently associated with perivascular infiltration by mononuclear cells (Fig. 6B) .

Demonstration of demyelination in the presence of MHV-4 protective antibody suggested that the effects of antibody were different on neurons and oligodendrocytes. Since the ts 8 mutant of MHV-4 causes infection of oligodendrocytes, we evaluated the effect of protective antibody on the demyelinating disease induced by ts 8 virus. Two groups of seven mice each received either protective MAb 5B19.2 or no antibody, and both groups were challenged 1 day later with 10,000 PFU of ts 8 virus ic. Demyelination was found in 100% of the mice examined from each group suggesting that MAb 5B19.2 can protect neuronal cells from histopathologic disease, but does not similarly protect oligodendrocytes. ts mutant of MHV-JHM (Haspel et aL, 19'78) virus showed decreased encephalitogenicity while retaining the ability to induce CNS demyelination.

In the present study, we have investigated the influence on MHV-4 infection of passively transferred antibodies to single epitopes on the viral glycoproteins.

Our results demonstrate that specific monoclonal antibodies to the E2 glycoprotein block encephalitis and convert a normally lethal infection to nonfatal demyelinating disease. This sparing effect upon passive transfer is restricted to a subset of MAb which react with the A (E2) and B (E2) epitopes but is not a property of antibody to epitope C (E2) or of MAb to the MHV El or N polypeptides (Talbot et aL, 1934) . Further, although all of the protective antibodies examined neutralized virus infectivity in vitro, one MAb (4B11.6) directed against epitope C (E2) neutralized virus efficiently in vitro but showed no sparing effect in viva, thus the properties of in vitro neutralization and in tivo protection can be distinguished and appear to map to distinct subsets of epitopes on E2.

Protective antibody does not prevent infection of CNS cells by MHV. Further, replication was decreased but not completely blocked by antibody. Virus titers in the brains of mice treated with protective doses of antibody prior to infection were approximately 5% of those reached in control unprotected mice (Table 3) . Antibody was found to be effective in conferring protection when given as late as l-2 days after virus challenge (Fig. 2) . Thus, the sparing effect is both therapeutic in that the titer of virus and severity of pathology are lessened, and prophylactic because lethal encephalitis is prevented. It is unlikely that the absence of liver lesions in antibodyprotected mice is a significant factor in their survival since much larger doses (1099 PFU) of MHV-4 than used in these studies cause only subacute hepatitis when given intraperitoneally.

The mechanism by which antibody-mediated protection functions appear to be in blocking the spread of infection in neurons. It is apparent from the data in Table 3 that virus replicates in the brains of protected mice but to diminished levels relative to control mice. Only a few neurons bearing viral antigen could be demonstrated in these brains by immunofluorescence (data not shown). Further, no evidence of MHV replication was found in neurons in the spinal cord. The sparing effect of MAb was selective for neuronal cells. Infected oligodendrocytes and resultant demyelination were observed in spinal cords of infected and passively protected mice (Figs. 5, 6) . Further, antibody which protected against lethal encephalitis did not block demyelination induced by the ts 8 mutant of MHV-4. The observed restriction of virus replication in the neuronal cells but not oligodendrocytes suggests that infection spreads via different mechanisms in these two cell types. Whether this is due to differences in the mode of cell-to-cell spread or reflects distinct receptors for MHV-4 on each cell type is a matter for further study. In addition to the altered virus tropism in the presence of antibody, we found that the characteristics of the demyelinating lesions were changed. Lesions in antibodyprotected mice showed increased cellularity with frequent perivascular cuffing in contrast to the lesions usually observed following MHV-4 infection.

This model system demonstrates that antibody response to precisely defined regions on a viral glycoprotein may induce profound changes in the pathogenesis of infection and the phenotype of disease. Conceptually, such factors may be of importance following primary infections, where an early response to a critical epitape(s) might influence the course of disease. Stohlman and Weiner (1981) have shown that mice surviving acute MHV-4 encephalomyelitis have neutralizing antibody in the circulation and develop chronic demyelination.

Similarly, the repertoire of response to a primary infection may influence the course of subsequent infection by a related virus. Human viral diseases such as subacute sclerosing panencephalitis (Vandvik, 1973) and progressive rubella panencephalitis (Wolinsky et al, 1976) occur in the presence of high titers of antibody to measles and rubella viruses, respectively. In the human demyelinating disease multiple sclerosis, there are elevated titers of antibodies to a number of common viruses such as measles, varicella, herpes, and others (Norrby, 19'78) . Although causal relationships between the respective viruses, the presence of these antibodies, and demyelinating disease have not been established, the present model system provides new insight into the po-tential role of selected specific antibodies in leading to host survival from fatal infection, with development of a chronic disease. Finally, this model offers the opportunity to study the factors which govern the infection of and spread within specific populations of CNS cells by MHV-4 at the molecular level.

A murine virus (JHM) causing disseminated encephalomyelitis with extensive destruction of myelin II pathology

Monoclonal antibodies to murine hepatitis virus-4 (strain JHM) define the viral glycoproteins responsible for attachment and cell-cell fusion

Temperature sensitive mutants of mouse hepatitis virus produce a high incidence of demyelination

Analysis of the functions of coronavirus glycoproteins by differential, inhibition of synthesis with Tunicamycin

Mouse hepatitis virus type-4 (JHM strain)-induced fatal central nervous system disease. I. Genetic control and the murine neuron as the susceptible site of disease

Selective localization of wild type and mutant mouse hepatitis virus (JHM strain) antigens in CNS tissues by fluorescence, light and electron microscopy

Selected mutants of mouse hepatitis virus type 4 (JHM strain) induce different CNS diseases

Mechanism of demyelination in JHM virus encephalomyelitis. Electron microscopic studies

Viral antibodies in multiple sclerosis

Chronic central nervous system demyelination in mice after JHM virus infection

Isolation of coronavirus envelope glycoproteins and interaction with the viral nucleocapsid

Topographical mapping of epitopes on the glycoproteins of murine hepatitis virus-4 (strain JHM): Correlation with biological activities

Immunopathological aspects in the patbogenesis of subacute sclerosing panencephalitis, with special references to the significance of the immune response in the central nervous system

Infectious leukoencephalitis: A critical comparison of certain experimental and naturally-occurring viral leukoencephalitides with experimental allergic encephalomyelitis

Structural polypeptides of the murine coronavirus JHM

Pathogenesis of demyelination induced by mouse hepatitis virus (JHM virus)

Progressive rubella panencephalitis. Arch. New-01