L* Protein of Theiler's Murine Encephalomyelitis Virus Is Required for Virus Growth in a Murine Macrophage-Like Cell Line

Abstract

We sought to confirm the importance of L* protein for growth of Theiler's murine encephalomyelitis virus (TMEV) in a macrophage-like cell line, J774-1. The protein is out of frame with the polyprotein and synthesized in DA but not GDVII subgroup strains of TMEV. A recombinant virus, DANCL*/GD, which substitutes the DA 5′ noncoding and L* coding regions for the corresponding regions of GDVII and synthesizes L* protein, grew with little restriction in J774-1 cells. In contrast, another recombinant virus, DANCL*-1/GD, which has an ACG rather than an AUG as the starting codon of L* protein at nucleotide 1079, resulting in no synthesis of L* protein, did not grow well. No significant difference between the rates of adsorption to J774-1 cells of these viruses was observed. RNase protection assay demonstrated that DANCL*/GD viral RNA significantly increased, whereas only a minimal increase was observed for DANCL*-1/GD. The present study suggests that L* protein is required for virus growth in macrophages.

Strains of Theiler's murine encephalomyelitis virus (TMEV), a member of the genus Cardiovirus of the family Picornaviridae, are divided into two subgroups on the basis of their different in vivo activities (13, 16, 18). GDVII subgroup strains cause acute polioencephalomyelitis in mice, with neither virus persistence nor demyelination. In contrast, the DA or TO subgroup strain induces an early nonfatal polioencephalomyelitis in weanling mice, followed by chronic demyelination with virus persistence. In vitro biological differences between the two subgroups are also reported; they include the size of plaques (10) and the crystalline arrays of the GDVII subgroup versus the membranous structures of the DA subgroup in infected BHK-21 cells (5). Recently, members of our group reported another in vitro biological activity distinguishing the two subgroups, i.e., virus growth in a murine macrophage cell line, J774-1 (15). The DA strain replicates in J774-1 cells, while the GDVII strain does not. This observation is of interest because macrophages are a major site of viral persistence, which is essential for demyelination (2, 11). Additional studies suggested that a 17-kDa viral protein, called L* protein, is responsible for virus growth in J774-1 cells (25).

L* protein is translated out of frame with the polyprotein from an alternative AUG, 13 nucleotides (nt) downstream from the authentic polyprotein AUG (8). L* protein is only synthesized in DA subgroup strains, since the L* AUG is present in DA but not in GDVII subgroup strains (14). A role for L* protein in TMEV-induced demyelination has been studied in loss-of-function experiments using DAL*-1 mutant virus (3, 7, 9, 25), which has a mutation of the AUG to ACG at nt 1079 in the background of a DA strain and, therefore, does not synthesize L* protein. In this study, we constructed chimeric viruses in which L* protein is synthesized in the background of the parental GDVII strain, in order to further explore the function of L* protein in virus growth in J774-1 cells.

A series of parental and chimeric cDNAs are shown in Fig. 1. pDAFL3 and pGDVIIFL2 are full-length infectious cDNA clones generated from DA and GDVII strains of TMEV, respectively (6, 23). pDAL*-1 is a mutant construct in which the only difference from pDAFL3 is a change of the AUG at nt 1079 to ACG (8). In order to be able to transfer the entire DA L* coding region to GDVII, the KpnI(nt 935)-NcoI(nt 1962) fragments of pDAFL3, pDAL*-1, and pGDVIIFL2 were subcloned separately and then an NheI site was introduced at nt 1545 by site-directed mutagenesis. pDANCL*/GD was constructed by the replacement of a pDAFL3 fragment (nt 1 to 1549) with the corresponding fragment of pGDVIIFL2. pDANCL*-1/GD was constructed by the replacement of nt 1 to 1549 pDAL*-1 fragment with the corresponding fragment of pGDVIIFL2.

FIG. 1.

Parental and recombinant cDNAs generated to delineate the role of L* protein for virus growth in J774-1 cells. The positions of the TMEV noncoding and coding areas are shown at the top. The DA genome and segments from it are shown as open bars, and the GDVII genome is shown as closed bars. The stars indicate point mutations of U to C in the L* starting codon at nt 1080. NheI indicates an NheI site introduced at the nt 1547 L* stop codon.

At first, to confirm the synthesis of L* protein in recombinant viruses in vivo, we radiolabeled BHK-21 cells infected with those recombinants. Radiolabeling of BHK-21 cells, which are permissive for TMEV and maintained in Eagle's minimum essential medium supplemented with 5% calf serum and 60 μg of kanamycin per ml, was performed as previously described (7). A quantity of 10⁶ cells in a 35-mm-diameter plastic culture dish were mock infected or infected with parental and recombinant viruses at a multiplicity of infection (MOI) of 10 PFU per cell. After adsorption at 37°C for 1 h, the cells were incubated in culture medium containing 2% calf serum and actinomycin D (2.4 μg/ml). After 9 h, the medium was replaced with methionine-free Eagle's minimum essential medium containing 1% calf serum and incubated for another 1 h. A total of 50 μCi of l-[³⁵S]methionine was then added. After 12 h, the labeled cells were scraped, washed twice with cold phosphate-buffered saline, and dissolved in the sample buffer (50 mM Tris-HCl [pH 6.8], 2% sodium dodecyl sulfate [SDS], 0.005% bromophenol blue, 10% glycerol, and 6% β-mercaptoethanol). Radiolabeled proteins were separated by SDS–15% polyacrylamide gel electrophoresis (PAGE) and analyzed by autoradiography. As shown in Fig. 2A, lanes 1 and 3, respectively, the parental DA and DANCL*/GD viruses, both of which have an L* coding sequence, produced a significant amount of 17-kDa protein. It was confirmed that the recombinant DANCL*/GD could produce almost the same amount of 17-kDa protein as the parental DA virus. On the other hand, as expected, no 17-kDa protein was detected following infection with the parental GDVII virus or the DANCL*-1/GD virus (Fig. 2A, lanes 2 and 4), since both viruses have an ACG rather than an AUG starting codon at nt 1079. In order to confirm whether the 17-kDa protein is L*, Western blotting was performed with a rabbit polyclonal antibody to synthetic peptides corresponding to L* protein amino acid residues 70 to 88 (NPRETPLHLTRVTPSPQVT). The peptides were synthesized and coupled to keyhole limpet hemocyanin by Sawady Technology Co., Inc. (Tokyo, Japan); the peptides were 92.4% pure as determined by high-pressure liquid chromatography. Unlabeled proteins were separated by SDS–15% PAGE and transferred to a nitrocellulose membrane (Amersham Pharmacia Biotech, Ltd., Little Chalfont, Buckinghamshire, United Kingdom). Horseradish peroxidase-linked anti-rabbit immunoglobulin G (Amersham Pharmacia Biotech, Ltd.) as the second antibody was detected by enhanced chemiluminescence reagents (Amersham Pharmacia Biotech, Ltd.). As shown in Fig. 2B, lanes 1 and 3, respectively, a 17-kDa band was bound to the anti-L* antibody in DA- and DANCL*/GD-infected cells.

FIG. 2.

L* protein synthesis of the parental and the recombinant viruses. (A) BHK-21 cells infected with virus at an MOI of 10 PFU per cell were radiolabeled with l-[³⁵S]methionine as described in the text. Synthesized proteins were analyzed by SDS–15% PAGE. (B) Electrophoresed proteins were analyzed by Western blotting with a rabbit polyclonal antibody to a synthetic peptide and horseradish peroxidase-linked anti-rabbit immunoglobulin G as the first and the second antibodies, respectively. In both SDS-PAGE and Western blotting analyses, a distinct 17-kDa band was clearly demonstrated in DA- and DANCL*/GD-infected cells, but not in GDVII-, DANCL*-1/GD-, and mock-infected cells. Lanes: 1, DA-infected cells; 2, GDVII-infected cells; 3, DANCL*/GD-infected cells; 4, DANCL*-1/GD-infected cells; 5, mock-infected cells. The arrow in panel A indicates a 17-kDa protein.

Next, the kinetics of growth of the recombinant viruses in BHK-21 and J774-1 cells was examined as previously described (15, 25). J774-1, an H-2^d macrophage-like cell line derived from a tumor of a female BALB/c mouse (22), was obtained from the Cancer Cell Repository, Tohoku University, Sendai, Japan, and maintained in RPMI 1640 medium supplemented with 10% fetal bovine serum. In a 35-mm-diameter plastic culture dish, 10⁶ cells were infected at an MOI of 10 PFU per cell with each virus. The culture supernatants and cell lysates were harvested at the indicated times (0, 3, 6, 12, and 24 h after infection) and titrated by a standard plaque assay on BHK-21 cells. As shown in Fig. 3A, in BHK-21 cells, the recombinant viruses that respectively contained (DANCL*/GD) and lacked (DANCL*-1/GD) the L* coding sequence showed similar growth kinetics; the titers of both cell-free and cell-associated viruses reached a peak of 10⁸ PFU/ml at 12 h postinoculation (p.i.). The titers then gradually decreased. However, the growth kinetics following infection of J774-1 cells differed for the two recombinant viruses (Fig. 3B). In DANCL*/GD infection, the titer peaked at 12 h p.i., at a level lower (by a multiple of 1 to 2 log units) than that seen in BHK-21 cells, and then decreased gradually. The data indicated that DANCL*/GD, which contains the L* AUG and coding sequence, can grow in J774-1 cells, although it is somewhat restricted compared to BHK-21 cells. On the other hand, the multiplication of DANCL*-1/GD without the L* AUG was severely restricted. As for DANCL*/GD, the titer reached a peak at 12 h p.i.; however, the level of virus was overall significantly lower, being at the most 10⁵ PFU/ml.

FIG. 3.

Kinetics of recombinant DANCL*/GD and DANCL*-1/GD viruses in BHK-21 (A) and J774-1 (B) cells. The culture supernatants and cell lysates of infected cells were harvested at the indicated times and subjected to titer determination by a standard plaque assay on BHK-21 cells. Data are the means of three independent experiments. Details of the construction are provided in the text and shown in Fig. 1. □, culture supernatant of DANCL*/GD-infected cells; ■, cell lysate of DANCL*/GD-infected cells; ○, culture supernatant of DANCL*-1/GD-infected cells; ●, cell lysate of DANCL*-1/GD-infected cells.

We next examined which step of the infection cycle restricted DANCL*-1/GD virus growth. The initial step of infection, i.e., adsorption of virus to J774-1 cells, was studied first. A quantity of 10⁶ cells in a 35-mm-diameter plastic culture dish was inoculated with each virus at an MOI of 10 PFU per cell. After virus adsorption at 4°C for 60 min, the supernatants were harvested and titrated by a standard plaque assay on BHK-21 cells. The rate of viral attachment to cells was calculated by the ratio of cell-attached virus titer to input virus titer according to the method described in a previous report (24). Statistical significance of the difference in the adsorption rates of DANCL*/GD and DANCL*-1/GD viruses was determined by a two-sample Student's t test. The adsorption rates of DANCL*/GD and DANCL*-1/GD viruses were 65.0 and 67.1%, respectively (data are means of three independent experiments). Statistical analysis showed no significant difference in the adsorption rates of these two viruses (P > 0.05).

Viral RNA synthesis of DANCL*/GD and DANCL*-1/GD viruses was finally examined by an RNase protection assay with J774-1 cells as previously described (25). pGDVIIFL2 was used as a template for in vitro transcription. To prepare an antisense probe, pGDVIIFL2 was cleaved at the PpuMI (nt 7575) site in the viral genome. The linearized plasmid was transcribed in vitro with T3 RNA polymerase in the presence of [α-³²P]UTP. The radiolabeled 577-nt-long RNA probe contained 526 nt of sequence complementary to the 3′ region of the viral genome and 51 nt of vector sequence (6, 19, 20). Total RNA was extracted and purified from 10⁶ virus-infected cells. A 2-μg portion of RNA and 100,000 cpm of [α-³²P]UTP-labeled probe were hybridized for 16 h at 42°C and treated with a mixture of 250 U of RNase A per ml and 10,000 U of RNase T₁ (Ambion, Inc., Austin, Tex.) per ml. The RNase-resistant fragments were detected by electrophoresis on a 5% polyacrylamide gel containing 8 M urea. The gels were exposed to X-ray film (Fuji Photo Film Co., Ltd., Kanagawa, Japan). As shown in Fig. 4, a significant amount of DANCL*/GD viral genomic RNA was detected at 3 h p.i., and it had clearly increased by 9 h p.i. On the other hand, DANCL*-1/GD viral genomic RNA was scarcely visible at 3 h p.i., with only a low amount detectable at 9 h p.i. The results were consistent with the growth kinetics of the recombinant viruses in J774-1 cells.

FIG. 4.

Genomic RNA synthesis in the recombinant virus-infected J774-1 cells. Total RNA was extracted from the cells of DANCL*/GD or DANCL*-1/GD virus at 3 and 9 h p.i. A 2-μg quantity of RNA was hybridized with [α-³²P]UTP-labeled riboprobe and treated with RNase solution as described in the text. The RNase-resistant fragments were denatured and electrophoresed on a 5% polyacrylamide–8 M urea gel. Lanes P and 1, [α-³²P]UTP-labeled probe and mock-infected J774-1 cells, respectively; lanes 2 and 3, J774-1 cells infected with DANCL*/GD at 3 and 9 h p.i., respectively; lanes 4 and 5, J774-1 cells infected with DANCL*-1/GD at 3 and 9 h p.i., respectively.

One of our goals is to delineate molecular determinants for TMEV growth in macrophages, which are a major site of virus persistence (4, 12). Our previous data suggested that L* protein, which is synthesized out of frame with the polyprotein from an alternative initiation codon in the L coding region of DA subgroup strains, is a key determinant of the cell type-specific restriction of virus growth (17). The previous study made use of a DA mutant virus which failed to synthesize L* protein. In the present study, in order to further confirm a role for the L* protein, we used a GDVII recombinant virus that contains the L* AUG and coding sequence. We substituted the L* coding region into GDVII, since the amino acid sequence homology of this region is only 79% for DA and GDVII strains (19, 21).

By comparing the activities of DANCL*/GD virus with DANCL*-1/GD virus, we confirmed the importance of the L* protein for virus growth in J774-1 cells, which importance was previously supported by studies with DAL*-1 mutant virus (25); DANCL*/GD with L* protein grew well in J774-1 cells, while DANCL*-1/GD without L* protein did not. Although 19% of the DA sequence (nt 1 to 1549) was substituted into GDVII to generate the recombinant viruses, the only difference between DANCL*/GD and DANCL*-1/GD in the chimeric segment was a U versus C at nt 1080. In addition, an RNase protection assay demonstrated the difference in the amount of viral genomic RNA between the two recombinant viruses. It remains to be elucidated which step regulates virus growth in J774-1 cells, i.e., polyprotein translation, polyprotein processing, or viral RNA replication.

The present investigation demonstrated that the two recombinant viruses showed similar growth kinetics in BHK-21 cells but different growth kinetics in J774-1 cells. This suggests that L* protein-dependent virus growth is host cell restricted. Our data also demonstrated that DA but not GDVII strains grow in particular monocyte or macrophage lineage cell lines but that both strain types grow equally in other neural and nonneural cell lines (17). The ability of DA to grow in macrophages is important for the maintenance of viral genome, which is one of the factors influencing the establishment of virus persistence (1). Therefore, the L* protein-dependent virus growth may be important in regulating DA-induced demyelination. The host cell restriction suggests that some host factor(s) interacts with the L* protein in influencing virus growth in macrophages. A clarification of the host factor and the mechanisms that underlie L* protein-dependent virus growth may lead to a better understanding of the basis for virus persistence in this system as well as in other viruses.