Persistent Theiler's Murine Encephalomyelitis Virus Infection in Mice depends on Plaque Size

J. gen. Virol. (198o), 46, 169-177 Printed in Great Britain 169 Persistent Theiler's Murine Encephalomyelitis Virus Infection in Mice depends on Plaque Size By H. L. LIPTON Department of Neurology, Northwestern University Medical School, 3o3 E. Chicago Avenue, Chicago, Illinois 60611, U.S.A. (Accepted 24 August I979) SUMMARY Theiler's murine encephalomyelitis virus (TMEV) is an enteric pathogen of mice which causes acute and chronic neurological disorders in the natural host. When brain-derived stocks of TMEV isolates are adapted to cell culture they predominantly form either large or small plaques. In this study the type of central nervous system (CNS) infection (acute versus chronic) and the associated disease occurring in mice inoculated intracerebrally with large and small plaque strains of TMEV was investigated. Large and small plaque strains of TMEV were found to vary in virulence, type of neurological disease produced and ability to establish persistent CNS infection in mice. Two large plaque strains, GDVII and FA viruses, were highly virulent, produced acute encephalitis, but were cleared from the nervous systems of surviving animals. Therefore, it appears that these large plaque variants do not cause persistent CNS infection in mice. In contrast, five small plaque strains, DA, WW, TO4, Yale and BeAn8386 viruses, were relatively avirulent, usually produced no illness during the first month after inoculation, but readily established persistent CNS infection in mice. Persistently infected mice later developed demyelinating disease. Having identified strains of TMEV that differ regarding their ability to persist, we now hope to be able to exploit this difference in elucidating the basic mechanism(s) of TMEV persistence. INTRODUCTION Theiler's murine encephalomyelitis virus (TMEV) is a natural enteric pathogen of mice which has been shown experimentally to produce persistent central nervous system (CNS) infection (Theiler, 1937; Lipton, 1975). In the first month after intracerebral (i.c.) inoculation of brain-derived virus stocks, virus replication and poliomyelitis-like pathological changes take place in neuronal populations in CNS grey matter (early disease) (Lipton, 1975). After this time surviving animals develop persistent CNS infection, but the cell type(s) in the nervous system in which this virus replicates remains to be elucidated. As a consequence, these chronically infected mice develop mononuclear cell infiltrates in their leptomeninges and white matter and develop lesions of primary demyelination (Dal Canto & Lipton, 1975). By using tissue culture adapted TMEV stocks, the small plaque variants were found to produce essentially late disease without antecedent early disease (Lipton, I978a; Lipton & Dal Canto, i979). The present study characterizes the type of CNS infection (acute versus chronic) and the associated disease with large and small plaque strains of TMEV. A preliminary report of oo2z-i317/8o/oooo-3785 $oz.oo 198o SGM

170 H.L. LIPTON this work was presented at the ICN-UCLA Symposium on Persistent Viruses, in Keystone, Colorado, U.S.A. (Lipton, I978b ). METHODS Animals and animal inoculations. Outbred, Swiss male mice (CD-I) were purchased from Charles River Breeding Laboratories (Portage, Mich.). All animals were inoculated into the right cerebral hemisphere with virus. Titrations of virus stocks were performed using 6 to Io mice per dilution, and end points were calculated by the method of Reed & Muench 0938). Cells and media. BHK2t cells were cultured in Dulbecco's modified Eagle medium (DME, Grand Island Biological Co., Grand Island, N.Y.) supplemented with o.i mm-lglutamine, IOO #g streptomycin, Ioo units penicillin per ml and Io% foetal calf serum. The same medium containing 2 % foetal calf serum was used for cell maintenance. Viruses. GDVII, FA, DA, TO4, WW, Yale, and BeAn8386 viruses were isolated in cell culture from brain-derived pools as described (Lipton, I978a ). These viruses were plaquepurified one to three times and stocks were prepared in BHK2I cells after several more passages to amplify the virus titres. Virus assays. The virus content of clarified CNS homogenates was determined by standard plaque assay as described (Rabinowitz & Lipton, I976). In brief, confluent monolayers of BHK2t cells in 35 mm plastic FB-6TC multiculture dishes (Linbro Chemical Co., Hamden, Conn.) were washed once with phosphate-buffered saline, ph 7"4 and inoculated with appropriate virus dilutions (o" I ml/well). After adsorption at 24 C for 60 min, the inoculum was removed and each well was overlaid with 3 ml of medium consisting of 0"95 % Noble agar (Difco Co., Detroit, Mich.) in DME containing 0"5 % bovine plasma albumin, fraction V (Armour Pharmaceutical Co., Chicago, II1.) and o. t mm-l-glutamine, ~oo/zg streptomycin, and Ioo units penicillin per ml. On the third day of incubation at 37 C, 1.5 ml of a second overlay containing o.oi % neutral red was added to each well and plaques were read 8 to 24 h later. Histology. Anaesthetized mice were sacrificed by total body perfusion with chilled 3 % glutaraldehyde in phosphate buffer, ph 7"4. Brains and spinal cords were fixed, embedded in paraffin or Epon and stained as described previously (Dal Canto & Lipton, I975). Neutralizing antibody. Twofold dilutions of heat-inactivated (56 C for 30 min) serum were incubated with 50 to IOO p.f.u, of GDVII virus at 24 C for 60 min. Each mixture was assayed in duplicate on BHKzl cell monolayers and neutralizing antibody was considered present if there was a 5o % reduction in p.f.u, compared with a virus-diluent control. RESULTS Plaque size and virulence Two of seven TMEV strains, GDVII and FA viruses, were found to form large plaques in BHK cells, whereas DA, TO4, WW, Yale and BeAn8386 viruses produced small plaques. To determine the neurovirulence, 4-week-old mice were inoculated i.c. with tenfold dilutions of these viruses. The mice were observed for clinical disease and CNS histopathological changes were studied in representative animals. GDVII and FA viruses regularly caused a fatal encephalitis while all of the viruses which formed small plaques largely produced asymptomatic infections during the first month (Table I). However, the small plaque viruses all persisted in the CNS (see below) and mice later developed prominent spinal cord leptomeningeal and white matter lesions consisting of mononuclear cell inflammation and

Persistent Theiler's murine virus infection I71 Table I. Correlation of disease state with plaque size for the Theiler's mouse encephalomyelitis viruses Virus Plaque size Disease state* log10 LDs0t GDVII Large Acute encephalitis I'l 3 FA Large Acute encephalitis I "04 DA Small Chronic demyelination > 5"79 TO4 Small Chronic demyelination > 4"0 WW Yale Small Small Chronic demyelination Chronic demyelination > 4"0 ND:I: BeAn8386 Small Chronic demyelination > 4"38 * At least five animals were examined histopathologically for each strain of virus. t Number of p.f.u, per LD~0. $ Not done. I I I I 8 7 6 Z 5 ~3 o -- I ill I I I 3 5 7 14 21 28 Time after infection (days) J.z I 35 / 70' ~- 105 Fig. i. Comparison of the growth of virus in the CNS of 4-week-old mice inoculated with IO 5 p.f.u. of GDVII virus (O-- ) and DA virus (O--Q). Each vertical bar represents the range of titres for three to four specimens with the curve intersecting at the mean value. Negative specimens are indicated by a downward pointing arrow orl the vertical bar. The level of sensitivity of the plaque assay was 5o p.f.u./g CNS tissue. m primary demyelination. These pathological changes were similar to the lesions described in detail for DA virus (Dal Canto & Lipton, i975). The virulence of the seven TMEV strains represented in number of p.f.u./lds 0 is shown in Table i. The large plaque viruses, GDVII and FA, were at least IOOO to 60000 times more virulent than the small plaque ones. Therefore, TMEV plaque size does appear to correlate with virulence. It is also apparent that the disease state produced by each plaque type differs. CNS virus replication The temporal course of virus replication in the CNS of mice was determined for a virulent (GDVII virus) and a relatively avirulent (DA virus) strain of TMEV (Fig. I). In both instances there was a rapid increase in virus content with maximum titres reached on days 5 to 7, thereafter followed by a decline in virus titres. GDVII virus levels were approx. IOOO higher than those observed for DA virus and GDVII virus-infected mice died by days 5 to 9. DA virus-infected mice remained well during this period and although there was a continual decline in virus content after day I4, infectious virus was detected in the CNS as late as 1o5 days after inoculation (the latest times tested in this experiment). To determine if there was a change in the virus during its replication in the CNS, the diam. of virus plaques from CNS were measured. As shown in Table 2, the CNS of GDVII

172 H. L. LIPTON Table 2. Size of Theiler's mouse encephalomyelitis virus plaques isolated from the CNS of mice Virus Day p.i. Mean plaque size s.e.* GDVII 5 2'47 +- 0"098 mm DA 7 < o'3 mm 2I < 0"3 113111 35 < 0"3 mm * Plaque size was determined by measuring the diam. of 5o plaques from each of three to five animals sacrificed on the days indicated. GDVII virus plaques ranged in size from i to 5 mm. Fig. z. Size of TMEV plaques isolated from mice infected with GDVII virus (left frame) and DA virus (right frame). virus-infected mice contained large plaque virus, whereas DA virus-infected mice yielded small plaque virus. The contrast in plaque size is shown in Fig. 2. Therefore, it would appear that TMEV replication in vivo is relatively stable without evident reversion of virus from one plaque type to the other. Persistent CNS infection initiated by other small plaque TMEV strains From the decline in the level of DA virus in the CNS of mice after days 7 to r 4 (Fig. 0, and since there are normal host immune responses to this agent (Rabinowitz & Lipton, I976), one would anticipate clearance of this CNS infection by day 2I. Therefore, mice having virus present after this time can be considered to be persistently infected. To determine if the other small plaque strains of TMEV cause persistent infection, mice inoculated with the TO4, WW, Yale and BeAn8386 strains of TMEV were sacrificed on day 28 for

Persistent Theiler's murine virus infection I73 Table 3. CNS virus content in mice inoculated intracerebrally with small plaque strains of Theiler's mouse encephalomyelitis virus loglo* Virus No. mice (mean virus titre) WW 4 I o ~'s6 Yale 8 lo 3's~ TO4 BeAn8386 4 6 1 oz'~s l o 4~3 * Mice were sacrificed for virus assay at 28 days p.i. Table 4. Examination of mice surviving GDVII virus titration for persbtent CNS infection Virus dose* Day sacrificed No. positive/ Expt. no. LDs0 p.lu. (p.i.) No. testedt 1 0"3 2 28 0/4 32"0 15o 28 I/I 2 0-6 8 21 0/5 ['3 15 Z1 0/3 I3"0 150 2I 0/I 3 0"8 15 21 0/4 1"3 24 2I 0/3 4 0"8 I5 22 0/6 I'3 24 21 0/5 2"0 38 2I 0/3 5 I'6 24 2I 0/4 IO-O 150 29 O/I * The amount of virus inoculated i.c. in number of 50 % median lethal doses and p.f.u. t Numerator represents the number of mice in which virus was detected and the denominator the total number tested. In experiment I only brains were tested; in the remaining experiments both the brain and spinal cord from each animal were assayed for virus. virus assay. Virus was detected in the spinal cords of mice inoculated with the other small plaque strains (Table 3). Furthermore, virus was also consistently found in the CNS of mice at later times after inoculation and the virus recovered was always small plaque in type (data not shown). Therefore, it would appear that the small plaque strains of TMEV are all capable of establishing persistent CNS infection in mice. Duration of CNS infection with large plaque TMEV Although the large plaque strains of TMEV are highly virulent, it is possible that they also cause persistent infection. In order to determine if the large plaque strains persist, mice surviving i.e. titration with GDVII virus were sacrificed for virus assay. The results for animals receiving o. 3 to 3 z LDs0 of virus are shown in Table 4. Infectious virus was detected in only one of 4o survivors and this animal was from the first titration. Subsequently, no other mice were found with persistent infection (Table 4)- It is possible that mice receiving these relatively low doses of a highly virulent virus, in fact never became infected. For this reason, neutralizing antibody titres were determined. Since TMEV is enterically transmitted, uninfected mice caged with i.p. inoculated mice were tested for antibody to exclude horizontal spread of infection leading to humoral immunity by days 2I to 28. The uninfected cage controls had antibody titres of r : 8 or less (data not shown). As shown in Table 5, in comparison to control infected mice which were inoculated i.p. only those mice receiving 2 to 13 LD50 of virus i.c. had as high a mean antibody titre. However, 15//26 of the others had a titre of 1 : 16 or more. Therefore, infection probably did take place in some animals

I74 H. L. LIPTON Table 5. Neutralizing antibody response of surviving GD VI1 virus-inoculated mice oti days 21: to 28 LDao virus dose No. mice Antibody titre* 0.2--0.8 I5 I4 (4-64) I-3-t'6 I I i2 (4-16) 2"o-I 3-o 4 40 (I6-64) Control infectedt 5 42 (8-128) * Reciprocal of mean antibody titre with the range in parentheses. t IO 3"~ p.f.u, of GDVII virus inoculated i.p. and mice sacrificed on day 2i. Table 6. Comparison of the LDso and IDso for large and small plaque strains of Theiler's mouse encephalomyelitis viruses Relative virulence Virus logxo (LD~o)* logio (1Dso)t (LD~o/ID6o) GDVII t'13 I'I3 I FA I '04 I "04 i DA BeAn8386 > 5'79 > 4"38 3"48 2"49 > I1o > ioo * The number of p.f.u./lds0. t The number of p.f.u./ids0. The IDa0 was determined by challenge of surviving mice at 60 to 90 days p.i. with 50 LDso of GDVII virus. Table 7- Failure of GDVII virus to cause persistent CNS infection after i.c. inoculation Virus assay* Challengedt [No. positive/no. (Protected/ Virus dose (Expt. no.) Survived/total tested (day)] total) 5 LDso I3/3o 0/8 (26-35) -- IO LDso (i) 3/40 0/3 (2I) -- (2) 16/102 0/6 (4 O) 8/IO~ * Brain and spinal cord of each animal was assayed for virus. The day p.i. mice were sacrificed is in parentheses. t Challenged i.c. with 5o LDs0 of GDVII virus on day 4o p.i. Mice were considered to be protected if they survived past day 7 since the median survival time of non-immune mice is 5 days. Median survival time was 9 days. inoculated with 0. 3 to 1.6 LDs0 of virus, but only regularly occurred in mice given 2 or more LDs0. Further evidence that non-lethal infection is not as readily induced with highly virulent TMEV is shown in Table 6. The median 5O~o infectious dose (IDs0) of four strains of TMEV was determined by challenging mice with 5o LDs0 of GDVII virus that had survived i.c. titration. A marked difference was demonstrated between the LDs0 and IDs0 for the two strains of small plaque TMEV. In contrast, the values were similar for the two strains of large plaque TMEV. Therefore the small plaque TMEV compared with their large plaque counterparts were relatively avirulent and it is more likely that an infection will result in host survival. Because of the difficulty in inducing non-lethal infections with low doses of GDVII virus, additional animals were inoculated with 5 to Io LDs0 of GDVII virus. As shown in Table 7, virus was not detected in the brains or spinal cords of 17 surviving mice. The fact that 8/Io mice surviving a io LDs0 dose showed prolonged survival after challenge i.c. with 50 LD~0 of GDVII virus indicates they had been infected and possessed some immunity (Table 7). Since host age may also be a factor in persistence (Nathanson, ~977), I4-week-old mice were inoculated i.c. with 5 LDs0 of GDVII virus. None of six surviving mice sacrificed

Persistent Theiler' s murine virus infection on day 25 had detectable virus in the CNS. Further evidence against the large plaque strains of TMEV producing persistent CNS infection came from our failure to demonstrate virus in the CNS of FA virus-infected mice surviving a dose of approx, z LDs0 (data not shown). Therefore, large plaque strains of TMEV do not appear to be able to persist in the CNS of mice. The one positive animal in Table 4 may represent contamination or laboratory error. I75 DISCUSSION The present study has demonstrated that large and small plaque strains of TMEV vary in virulencc, typc of ncurological disease produccd and ability to establish persistent CNS infection (Table I). Two large plaque strains, GDVII and FA viruses, were highly virulent and produced acute encephalitis, but wcre cleared from the nervous systems of surviving animals. In contrast, five small plaque strains, DA, WW, TO4, Yale and BeAn8386 viruses, were relatively avirulent, usually produced no illness during the first month after inoculation, but readily established persistent CNS infection in mice which ultimately resulted in demyelinating disease. A relationship between virus plaque size and virulence is well documented in the literature (Takemoto, I966). This association has been demonstrated for members of virtually all classes of animal viruses and although this association has not been absolute, plaque size has been one of the more common markers used to distinguish virulent from attenuated viruses. In general, large plaque variants have been more virulent than their small plaque counterparts, as has now been demonstrated for TMEV. A number of other picornaviruses, including poliovirus types I (Vogt et al. 1957) and 2 (Takemori et al. I957), Coxsackievirus A9 (Takemoto & Habel, I959) and Mengo virus (Colter et al. I964) have shown a simitar relationship between plaque size and virulence. A reason for this association may be that faster growing viruses are more likely to be lethal for animals and to produce larger foci of cell destruction in vitro. However, our knowledge of this relationship has not resulted in a better understanding of the molecular basis of virulence. This study has provided no further insight into this problem. A major finding of this study is that the small plaque strains of TMEV cause persistent CNS infection (Fig. I, Table 3), whereas the large plaque strains do not (Tables 4 and 6). While the small plaque strains are relatively avirulent (attenuated) and infections initiated by them are less likely to be lethal (Table 6), this in itself probably does not explain the observed difference in the ability of the two plaque types to persist. In other words attenuation would not seem to be the only prerequisite for a virus to be able to persist. This notion is supported by the experience gained with a number of live, attenuated virus vaccines, since none has been reported to cause persistent infection in an immunocompetent host. However, the fact that such live attenuated vaccine viruses may not reach and infect a target organ, where persistence can be established (e.g. poliovirus in brain), needs to be taken into consideration. Persistence is obviously a complex biological phenomenon involving virus-host as well as virus-cell interactions. There are a number of ways in which virus-cell interactions can be modulated so that persistence develops. These mechanisms include generation of ts mutants (Preble & Youngner, 1975), formation of defective-interfering (DI) particles (Huang & Baltimore, 1977), elaboration of virus-induced protein(s) which can inhibit homologous virus replication (ter Meulen & Martin, ~976; Riedel & Brown, I979) and induction of interferon. As yet, none of these mechanisms has been implicated in TMEV persistence. Replication of the small plaque strains of TMEV is not temperature sensitive so they cannot be viewed as ts mutants of their large plaque counterparts (H. L. Lipton, unpublished data).

I76 H.L. LIPTON It is interesting to note that while DI particles have been identified in most animal virus systems, the production of DI particles by picornaviruses may be a rare event. It is true that poliovirus DI particles have been identified by two different groups (Cole et al. I97I; McLaren & Holland, I974); however, many other investigators have failed to generate poliovirus DI particles upon repeated passage at high multiplicity (see Nomoto et al. I979). In addition, others have been unsuccessful in generating DI particles upon many repeated passages of Mengo virus (J. J. Holland, personal communication) and Coxsackievirus B3 (Gaunt et al. I979) at high multiplicity. Whether the production of DI particles by picornaviruses will turn out to be uncommon and therefore an unlikely mechanism of persistence should be resolved in the future. Having identified strains of TMEV that differ regarding persistence, we now hope to be able to exploit this difference in elucidating the basic mechanism(s) of TMEV persistence. This study was supported by United States Public Health Service Grant AI 14I 39. H. L. L. is the recipient of Public Health Service Career Development Award IKO4-Aloo228. I thank Kimiko K. Matsutani and Barbara Klement for excellent technical assistance and Jody Leimbach for preparation of this manuscript. REFERENCES COLE, C., SMOLER, D., WlMMER, E. & BALTIMORE, D. 0971)- Defective interfering particles of poliovirus. I. Isolation and physical properties. Journal of Virology 7, 478-485. COLTER, J. S., CAM~'BELL, J. a. & HATCH, L. R. (I964). The pathogenicity to mice of three variants of Mengo encephalomyelitis virus. Journal of Cellular and Comparative Physiology 65, 229-236. DAL CANTO, M. C. & LIVTON, H.L. 0975). Primary demyelination in Theiler's virus infection. An ultrastructural study. Laboratory Investigation 33, 626-637. GAUNT, C. J., TROUSDALE, M. D., LABADIE, O. R. & PAQUE, R. E. (1979). Lack of production of defective-interfering particles by Coxsackie-virus B3 in HeLa cells. Abstracts of the Annual Meeting of the American Society for Microbiology. Los Angeles, California. HUANG, A. S. & BALTIMORE, D. 0977)- Defective interfering animal viruses. In Comprehensive Virology, vol. IO, chapter 2, pp. 73-I i6. Edited by H. Fraenkel-Conrat and R. R. Wagner. New York: Plenum Publishing Corporation. LIPTON, H. L. (1975)- Theiler's virus infection in mice: an unusual biphasic disease process leading to demyelination. Infection & Immunity xx, 1147-I 155. LrPTON, H.L. (I978a). Characteristics of the TO strains of Theiler's mouse encephalomyelitis viruses. Infection & Immunity 2o, 869-872. LIPTON, H. L. (1978 b). The relationship of Theiler's mouse encephalomyelitis virus plaque size with persistent infection. In Persistent Viruses: 1CN-UCLA Symposia on Molecular and Cellular Biology, vol. II, pp. 679-689. Edited by J. Stevens, G. J. Todaro and C. F. Fox. New York: Academic Press. LIPTON, H. L. & OAL CANTO, M. C. (1979). The TO strains of Theiler's viruses cause 'slow virus-like' infections in mice. Annals of Neurology 6, 25--28. MCLAREN, L. C. & HOLLAND, J. J. (I974). Defective interfering particles from poliovirus vaccine and vaccine reference strains. Virology 60, 579-583. NATHANSON, N. (I977). Persistent viral infections of humans: an overview. In Microbiology, pp. 533-538. Edited by D. Schlessinger. Washington, D.C. : American Society for Microbiology. NOMOTO, A., JACOBSON, A., LEE, Y. F., DUNN, J. & WIMMER, E. (I979). Defective interfering particles of poliovirus: mapping of the deletion and evidence that the deletions in the genomes of DI (I), (2) and (3) are located in the same region. Journal of Molecular Biology 128, 179-196. PREBLE, O. T. & YOUNGNER, J. S. (1975)- Temperature-sensitive viruses and the etiology of chronic and inapparent infection. Journal of Infectious Diseases x3x, 467. RABINOW1TZ, S. G. & LIPTON, H. L. (1976). Cellular immunity in chronic Theiler's virus central nervous system infection. Journal of Immunology xx7, 357-363. REED, L J. & MUENCH, H. A. (1938). A simple method of estimating fifty percent endpoints. American Journal of Hygiene ~'7, 493-497. RIEDEL, B. & BROWN, D. T. (I979). Novel antiviral activity found in the media of Sindbis virus - persistently infected mosquito (Aedes albopictus) cell cultures. Journal of Virology 29, 5 I-6O. TAKEMORI, N., NOMURA, S., MORIOKA, M. M. & KITAOKA, M. (1957). Minute plaque mutant of type 2 poliovirus. Science I26, 924-925.

Persistent Theiler's murine virus infection I77 TAKEMOTO, K. K. (1966). Plaque mutants of animal viruses. Progress in Medical Virology 8, 314-348. TAKEMOTO, K. K. & HABEL, K. (1959). Virus-cell relationship in a carrier culture of HeLa cells and Coxsackievirus A9 virus. Virology 7, 28-44. TER MEULEN, V. & MARTIN, S. J. (1976). Genesis and maintenance of a persistent infection by canine distemper virus. Journal of General Virology 32, 43 t-44o. THEILER, M. (1937)- Spontaneous encephalomyelitis of mice, a new virus disease. Journal of Experimental Medicine 72, 79-83. VOGT, M., DULBECCO, a. & WENNER, H. A. (1957). Mutants of poliomyelitis viruses with reduced efficiency of plating in acid medium and reduced neuropathogenicity. Virology 4, 14I-I 55. (Received 24 May I979)